-
Schedule Amazon SageMaker notebook jobs and manage multi-step notebook workflows using APIsby Anchit Gupta (AWS Machine Learning Blog) on November 29, 2023
Amazon SageMaker Studio provides a fully managed solution for data scientists to interactively build, train, and deploy machine learning (ML) models. Amazon SageMaker notebook jobs allow data scientists to run their notebooks on demand or on a schedule with a few clicks in SageMaker Studio. With this launch, you can programmatically run notebooks as jobs
-
Announcing new tools and capabilities to enable responsible AI innovationby Peter Hallinan (AWS Machine Learning Blog) on November 29, 2023
The rapid growth of generative AI brings promising new innovation, and at the same time raises new challenges. These challenges include some that were common before generative AI, such as bias and explainability, and new ones unique to foundation models (FMs), including hallucination and toxicity. At AWS, we are committed to developing generative AI responsibly,
-
Introducing the AWS Generative AI Innovation Center’s Custom Model Program for Anthropic Claudeby Sri Elaprolu (AWS Machine Learning Blog) on November 29, 2023
Since launching in June 2023, the AWS Generative AI Innovation Center team of strategists, data scientists, machine learning (ML) engineers, and solutions architects have worked with hundreds of customers worldwide, and helped them ideate, prioritize, and build bespoke solutions that harness the power of generative AI. Customers worked closely with us to prioritize use cases,
-
A conversation with Refik Anadol on creativity and AIby (AI) on November 29, 2023
Mira Lane explores how AI is empowering creativity with pioneering artist Refik Anadol.
-
Millions of new materials discovered with deep learningby Google DeepMind Blog on November 29, 2023
We share the discovery of 2.2 million new crystals – equivalent to nearly 800 years’ worth of knowledge. We introduce Graph Networks for Materials Exploration (GNoME), our new deep learning tool that dramatically increases the speed and efficiency of discovery by predicting the stability of new materials.
-
The Power of Promptingby Alyssa Hughes (Microsoft Research) on November 29, 2023
Microsoft Chief Scientific Officer Eric Horvitz explains how new prompting strategies can enable generalist large language models like GPT-4 to achieve exceptional expertise in specific domains, such as medicine, and outperform fine-tuned specialist models. The post The Power of Prompting appeared first on Microsoft Research.
-
Learn how to assess the risk of AI systemsby Mia Mayer (AWS Machine Learning Blog) on November 28, 2023
Artificial intelligence (AI) is a rapidly evolving field with the potential to improve and transform many aspects of society. In 2023, the pace of adoption of AI technologies has accelerated further with the development of powerful foundation models (FMs) and a resulting advancement in generative AI capabilities. At Amazon, we have launched multiple generative AI
-
Embracing Transformation: AWS and NVIDIA Forge Ahead in Generative AI and Cloud Innovationby Brian Caulfield (NVIDIA Blog) on November 28, 2023
Amazon Web Services and NVIDIA will bring the latest generative AI technologies to enterprises worldwide. Combining AI and cloud computing, NVIDIA founder and CEO Jensen Huang joined AWS CEO Adam Selipsky Tuesday on stage at AWS re:Invent 2023 at the Venetian Expo Center in Las Vegas. Selipsky said he was “thrilled” to announce the expansion Read article >
-
NVIDIA BioNeMo Enables Generative AI for Drug Discovery on AWSby Kimberly Powell (NVIDIA Blog) on November 28, 2023
Researchers and developers at leading pharmaceutical and techbio companies can now easily deploy NVIDIA Clara software and services for accelerated healthcare through Amazon Web Services. Announced today at AWS re:Invent, the initiative gives healthcare and life sciences developers using AWS cloud resources the flexibility to integrate NVIDIA-accelerated offerings such as NVIDIA BioNeMo — a generative Read article >
-
NVIDIA GPUs on AWS to Offer 2x Simulation Leap in Omniverse Isaac Sim, Accelerating Smarter Robotsby Gerard Andrews (NVIDIA Blog) on November 28, 2023
Developing more intelligent robots in the cloud is about to get a speed multiplier. NVIDIA Isaac Sim and NVIDIA L40S GPUs are coming to Amazon Web Services, enabling developers to build and deploy accelerated robotics applications in the cloud. Isaac Sim, an extensible simulator for AI-enabled robots, is built on the NVIDIA Omniverse development platform Read article >
-
NVIDIA Powers Training for Some of the Largest Amazon Titan Foundation Modelsby Nirmalya De (NVIDIA Blog) on November 28, 2023
Everything about large language models is big — giant models train on massive datasets across thousands of NVIDIA GPUs. That can pose a lot of big challenges for companies pursuing generative AI. NVIDIA NeMo, a framework for building, customizing and running LLMs, helps overcome these challenges. A team of experienced scientists and developers at Amazon Read article >
-
3D Artist Nourhan Ismail Brings Isometric Innovation ‘In the NVIDIA Studio’ With Adobe After Effects and Blenderby Gerardo Delgado (NVIDIA Blog) on November 28, 2023
This week’s talented In the NVIDIA Studio artist, Nourhan Ismail, created a literal NVIDIA studio.
-
Monitoring medical AI device adoptionby Machine learning : nature.com subject feeds on November 28, 2023
-
Data-driven grading of acute graft-versus-host diseaseby Machine learning : nature.com subject feeds on November 28, 2023
-
Adjust the format of papers to improve description by AIby Machine learning : nature.com subject feeds on November 28, 2023
-
Introducing three new NVIDIA GPU-based Amazon EC2 instancesby Chetan Kapoor (AWS Machine Learning Blog) on November 27, 2023
Amazon Elastic Compute Cloud (Amazon EC2) accelerated computing portfolio offers the broadest choice of accelerators to power your artificial intelligence (AI), machine learning (ML), graphics, and high performance computing (HPC) workloads. We are excited to announce the expansion of this portfolio with three new instances featuring the latest NVIDIA GPUs: Amazon EC2 P5e instances powered
-
Boost inference performance for LLMs with new Amazon SageMaker containersby Michael Nguyen (AWS Machine Learning Blog) on November 27, 2023
Today, Amazon SageMaker launches a new version (0.25.0) of Large Model Inference (LMI) Deep Learning Containers (DLCs) and adds support for NVIDIA’s TensorRT-LLM Library. With these upgrades, you can effortlessly access state-of-the-art tooling to optimize large language models (LLMs) on SageMaker and achieve price-performance benefits – Amazon SageMaker LMI TensorRT-LLM DLC reduces latency by 33%
-
Simplify data prep for generative AI with Amazon SageMaker Data Wranglerby Ajjay Govindaram (AWS Machine Learning Blog) on November 27, 2023
Generative artificial intelligence (generative AI) models have demonstrated impressive capabilities in generating high-quality text, images, and other content. However, these models require massive amounts of clean, structured training data to reach their full potential. Most real-world data exists in unstructured formats like PDFs, which requires preprocessing before it can be used effectively. According to IDC,
-
Democratize ML on Salesforce Data Cloud with no-code Amazon SageMaker Canvasby Daryl Martis (AWS Machine Learning Blog) on November 27, 2023
This post is co-authored by Daryl Martis, Director of Product, Salesforce Einstein AI. This is the third post in a series discussing the integration of Salesforce Data Cloud and Amazon SageMaker. In Part 1 and Part 2, we show how the Salesforce Data Cloud and Einstein Studio integration with SageMaker allows businesses to access their
-
GPT-4’s potential in shaping the future of radiologyby Brenda Potts (Microsoft Research) on November 27, 2023
This research paper is being presented at the 2023 Conference on Empirical Methods in Natural Language Processing (opens in new tab) (EMNLP 2023), the premier conference on natural language processing and artificial intelligence. In recent years, AI has been increasingly integrated into healthcare, bringing about new areas of focus and priority, such as diagnostics, treatment The post GPT-4’s potential in shaping the future of radiology appeared first on Microsoft […]
-
AWS AI services enhanced with FM-powered capabilitiesby Bratin Saha (AWS Machine Learning Blog) on November 27, 2023
Artificial intelligence (AI) continues to transform how we do business and serve our customers. AWS offers a range of pre-trained AI services that provide ready-to-use intelligence for your applications. In this post, we explore the new AI service capabilities and how they are enhanced using foundation models (FMs). We focus on the following major updates
-
Elevate your self-service assistants with new generative AI features in Amazon Lexby Anuradha Durfee (AWS Machine Learning Blog) on November 27, 2023
In this post, we talk about how generative AI is changing the conversational AI industry by providing new customer and bot builder experiences, and the new features in Amazon Lex that take advantage of these advances. As the demand for conversational AI continues to grow, developers are seeking ways to enhance their chatbots with human-like
-
Amazon Transcribe announces a new speech foundation model-powered ASR system that expands support to over 100 languagesby Sumit Kumar (AWS Machine Learning Blog) on November 26, 2023
Amazon Transcribe is a fully managed automatic speech recognition (ASR) service that makes it straightforward for you to add speech-to-text capabilities to your applications. Today, we are happy to announce a next-generation multi-billion parameter speech foundation model-powered system that expands automatic speech recognition to over 100 languages. In this post, we discuss some of the
-
Drive hyper-personalized customer experiences with Amazon Personalize and generative AIby Jingwen Hu (AWS Machine Learning Blog) on November 26, 2023
Today, we are excited to announce three launches that will help you enhance personalized customer experiences using Amazon Personalize and generative AI. Whether you’re looking for a managed solution or build your own, you can use these new capabilities to power your journey. Amazon Personalize is a fully managed machine learning (ML) service that makes
-
Build brand loyalty by recommending actions to your users with Amazon Personalize Next Best Actionby Shreeya Sharma (AWS Machine Learning Blog) on November 26, 2023
Amazon Personalize is excited to announce the new Next Best Action (aws-next-best-action) recipe to help you determine the best actions to suggest to your individual users that will enable you to increase brand loyalty and conversion. Amazon Personalize is a fully managed machine learning (ML) service that makes it effortless for developers to deliver highly
-
Medical Imaging AI Made Easier: NVIDIA Offers MONAI as Hosted Cloud Serviceby Prerna Dogra (NVIDIA Blog) on November 26, 2023
NVIDIA today launched a cloud service for medical imaging AI to further streamline and accelerate the creation of ground-truth data and training of specialized AI models through fully managed, cloud-based application programming interfaces. NVIDIA MONAI cloud APIs — announced at the annual meeting of RSNA, the Radiological Society of North America, taking place this week Read article >
-
Accelerating AI/ML development at BMW Group with Amazon SageMaker Studioby Marc Neumann (AWS Machine Learning Blog) on November 24, 2023
This post is co-written with Marc Neumann, Amor Steinberg and Marinus Krommenhoek from BMW Group. The BMW Group – headquartered in Munich, Germany – is driven by 149,000 employees worldwide and manufactures in over 30 production and assembly facilities across 15 countries. Today, the BMW Group is the world’s leading manufacturer of premium automobiles and
-
Automating product description generation with Amazon Bedrockby Dhaval Shah (AWS Machine Learning Blog) on November 24, 2023
In today’s ever-evolving world of ecommerce, the influence of a compelling product description cannot be overstated. It can be the decisive factor that turns a potential visitor into a paying customer or sends them clicking off to a competitor’s site. The manual creation of these descriptions across a vast array of products is a labor-intensive
-
Optimizing costs for Amazon SageMaker Canvas with automatic shutdown of idle appsby Davide Gallitelli (AWS Machine Learning Blog) on November 24, 2023
Amazon SageMaker Canvas is a rich, no-code Machine Learning (ML) and Generative AI workspace that has allowed customers all over the world to more easily adopt ML technologies to solve old and new challenges thanks to its visual, no-code interface. It does so by covering the ML workflow end-to-end: whether you’re looking for powerful data
-
How SnapLogic built a text-to-pipeline application with Amazon Bedrock to translate business intent into actionby Greg Benson (AWS Machine Learning Blog) on November 24, 2023
This post was co-written with Greg Benson, Chief Scientist; Aaron Kesler, Sr. Product Manager; and Rich Dill, Enterprise Solutions Architect from SnapLogic. Many customers are building generative AI apps on Amazon Bedrock and Amazon CodeWhisperer to create code artifacts based on natural language. This use case highlights how large language models (LLMs) are able to
-
Hungry for Gaming: 18 New Games to Join GeForce NOWby GeForce NOW Community (NVIDIA Blog) on November 23, 2023
GeForce NOW is bringing 18 new games to the cloud this week, part of a gratitude-filled GFN Thursday. A collaboration between Chromebook Plus, CD PROJEKT RED and GeForce NOW brought an immersive 3D activation to Times Square over the weekend, containing a hidden Easter egg for Cyberpunk 2077 players. Plus, this holiday season, give the Read article >
-
Amazon EC2 DL2q instance for cost-efficient, high-performance AI inference is now generally availableby A K Roy (AWS Machine Learning Blog) on November 22, 2023
This is a guest post by A.K Roy from Qualcomm AI. Amazon Elastic Compute Cloud (Amazon EC2) DL2q instances, powered by Qualcomm AI 100 Standard accelerators, can be used to cost-efficiently deploy deep learning (DL) workloads in the cloud. They can also be used to develop and validate performance and accuracy of DL workloads that
-
Your guide to generative AI and ML at AWS re:Invent 2023by Denis V. Batalov (AWS Machine Learning Blog) on November 22, 2023
Yes, the AWS re:Invent season is upon us and as always, the place to be is Las Vegas! You marked your calendars, you booked your hotel, and you even purchased the airfare. Now all you need is some guidance on generative AI and machine learning (ML) sessions to attend at this twelfth edition of re:Invent. And although generative AI has appeared in previous events, this year we’re taking it to the next level. In addition to several exciting announcements during keynotes, […]
-
Research Focus: Week of November 22, 2023by Alyssa Hughes (Microsoft Research) on November 22, 2023
A new deep-learning compiler for dynamic sparsity; Tongue Tap could make tongue gestures viable for VR/AR headsets; Ranking LLM-Generated Loop Invariants for Program Verification; Assessing the limits of zero-shot foundation models in single-cell biology. The post Research Focus: Week of November 22, 2023 appeared first on Microsoft Research.
-
Teenage Dream: Aspiring Computer Science Major Experiences NVIDIA Life With Make-A-Wish Visitby Samantha Zee (NVIDIA Blog) on November 22, 2023
A calendar packed with meetings, calls and lab visits may sound like a typical workday for many — but for Luca Lofranco, whose greatest wish was to experience what it’s like to work at NVIDIA, it was a dream come true. Eighteen-year-old Lofranco recently traveled from his hometown near Toronto, Canada, to spend the day Read article >
-
Build a contextual chatbot for financial services using Amazon SageMaker JumpStart, Llama 2 and Amazon OpenSearch Serverless with Vector Engineby Sunil Padmanabhan (AWS Machine Learning Blog) on November 22, 2023
The financial service (FinServ) industry has unique generative AI requirements related to domain-specific data, data security, regulatory controls, and industry compliance standards. In addition, customers are looking for choices to select the most performant and cost-effective machine learning (ML) model and the ability to perform necessary customization (fine-tuning) to fit their business use cases. Amazon
-
AI-Powered Tech Company Helps Grocers Start Afresh in Supply Chain Managementby Kristen Yee (NVIDIA Blog) on November 22, 2023
Talk about going after low-hanging fruit. Afresh is an AI startup that helps grocery stores and retailers reduce food waste by making supply chains more efficient. In the latest episode of NVIDIA’s AI Podcast, host Noah Kravitz spoke with the company’s cofounder and president, Nathan Fenner, about its mission, offerings and the greater challenge of Read article >
-
NVIDIA Collaborates With Genentech to Accelerate Drug Discovery Using Generative AIby Rory Kelleher (NVIDIA Blog) on November 21, 2023
Genentech, a member of the Roche Group, is pioneering the use of generative AI to discover and develop new therapeutics and deliver treatments to patients more efficiently. A new collaboration between Genentech, the biotechnology pioneer, and NVIDIA aims to transform the discovery and development of new medicines by bringing together experts from each company to Read article >
-
3D Artist Cooks Up Stunningly Photorealistic Food Renders This Week ‘In the NVIDIA Studio’by Gerardo Delgado (NVIDIA Blog) on November 21, 2023
It’s the season of gratitude: that time of year to give thanks for the people and small moments that make life so special.
-
Orca 2: Teaching Small Language Models How to Reasonby Alyssa Hughes (Microsoft Research) on November 21, 2023
At Microsoft, we’re expanding AI capabilities by training small language models to achieve the kind of enhanced reasoning and comprehension typically found only in much larger models. The post Orca 2: Teaching Small Language Models How to Reason appeared first on Microsoft Research.
-
How AI is expanding art historyby Machine learning : nature.com subject feeds on November 21, 2023
-
Designing molecules with autoencoder networksby Machine learning : nature.com subject feeds on November 21, 2023
-
AI should focus on equity in pandemic preparednessby Machine learning : nature.com subject feeds on November 21, 2023
-
Lifelong model editing in large language models: Balancing low-cost targeted edits and catastrophic forgettingby Alyssa Hughes (Microsoft Research) on November 20, 2023
Lifelong model editing fixes mistakes discovered after model deployment. This work could expand sequential editing to model properties like fairness and privacy and enable a new class of solutions for adapting LLMs over long deployment lifetimes. The post Lifelong model editing in large language models: Balancing low-cost targeted edits and catastrophic forgetting appeared first on Microsoft Research.
-
Abstracts: November 20, 2023by Brenda Potts (Microsoft Research) on November 20, 2023
Today I'm talking to Shrey Jain, an applied scientist at Microsoft Research, and Dr. Zoë Hitzig, a junior fellow at the Harvard Society of Fellows. The post Abstracts: November 20, 2023 appeared first on Microsoft Research.
-
What Is a SuperNIC?by Itay Ozery (NVIDIA Blog) on November 20, 2023
Generative AI is the latest turn in the fast-changing digital landscape. One of the groundbreaking innovations making it possible is a relatively new term: SuperNIC. What Is a SuperNIC? SuperNIC is a new class of network accelerators designed to supercharge hyperscale AI workloads in Ethernet-based clouds. It provides lightning-fast network connectivity for GPU-to-GPU communication, achieving Read article >
-
Accelerating climate action with AIby (AI) on November 20, 2023
AI has the potential to mitigate 5-10% of global greenhouse gas emissions according to our new report with Boston Consulting Group.
-
Skeleton-of-Thought: Parallel decoding speeds up and improves LLM outputby Brenda Potts (Microsoft Research) on November 17, 2023
Large language models (LLMs) such as LLaMA and OpenAI’s GPT-4 are revolutionizing technology. However, one of the common complaints about LLMs is their speed, or lack thereof. In many cases, it takes a long time to get an answer from them. This limits LLMs’ applications and their usefulness in latency-critical functions, such as chatbots, copilots, The post Skeleton-of-Thought: Parallel decoding speeds up and improves LLM output appeared first on Microsoft Research.
-
From Guangzhou to Los Angeles, Automakers Dazzle With AI-Powered Vehiclesby Calisa Cole (NVIDIA Blog) on November 17, 2023
Good news for car lovers: Two acclaimed auto shows, taking place now through next week, are delighting attendees with displays of next-generation automotive designs powered by AI. Hundreds of thousands of auto enthusiasts worldwide are expected to visit Guangzhou, China — known as the city of flowers — to attend its auto show, running through Read article >
-
AI to See ‘Major Second Wave,’ NVIDIA CEO Says in Fireside Chat With iliad Group Execby Brian Caulfield (NVIDIA Blog) on November 17, 2023
European startups will get a massive boost from a new generation of AI infrastructure, NVIDIA founder and CEO Jensen Huang said Friday in a fireside chat with iliad Group Deputy CEO Aude Durand — and it’s coming just in time. “We’re now seeing a major second wave,” Huang said of the state of AI during Read article >
-
NVIDIA and Scaleway Speed Development for European Startups and Enterprisesby Serge Lemonde (NVIDIA Blog) on November 17, 2023
Europe’s startup ecosystem is getting a boost of accelerated computing for generative AI. NVIDIA and cloud service provider (CSP) Scaleway are working together to deliver access to GPUs, NVIDIA AI Enterprise software, and services for turbocharging large language models (LLMs) and generative AI development for European startups. Scaleway, a subsidiary of French telecommunications provider iliad Read article >
-
OpenAI announces leadership transitionby OpenAI Blog on November 17, 2023
-
AI Training AI: GatorTronGPT at the Forefront of University of Florida’s Medical AI Innovationsby Mona Flores (NVIDIA Blog) on November 16, 2023
How do you train an AI to understand clinical language with less clinical data? Train another AI to synthesize training data. Artificial intelligence is changing the way medicine is done, and is increasingly being used in all sorts of clinical tasks. This is fueled by generative AI and models like GatorTronGPT, a generative language model Read article >
-
Three Ways Generative AI Can Bolster Cybersecurityby David Reber Jr. (NVIDIA Blog) on November 16, 2023
Human analysts can no longer effectively defend against the increasing speed and complexity of cybersecurity attacks. The amount of data is simply too large to screen manually. Generative AI, the most transformative tool of our time, enables a kind of digital jiu jitsu. It lets companies shift the force of data that threatens to overwhelm Read article >
-
What’s Your Story: Desney Tanby Alyssa Hughes (Microsoft Research) on November 16, 2023
From service in the Singapore Armed Forces to autonomous navigation with NASA & VR with Disney, Desney Tan’s life journey hasn’t been linear. Learn how Tan landed at Microsoft & about the purpose guiding his work in the podcast series “What’s Your Story”. The post What’s Your Story: Desney Tan appeared first on Microsoft Research.
-
Into the Omniverse: OpenUSD Enhancements for Autodesk Maya Make 3D Workflows a Ferret-Taleby Gerardo Delgado (NVIDIA Blog) on November 16, 2023
3D artists can improve the productivity and efficiency of their generative AI-enabled content-creation workflows thanks to the latest updates to popular OpenUSD software.
-
New ways generative AI can help you find holiday giftsby (AI) on November 16, 2023
Holiday shoppers can use generative AI to find gifts for others and themselves.
-
Transforming the future of music creationby Google DeepMind Blog on November 16, 2023
Announcing our most advanced music generation model and two new AI experiments, designed to open a new playground for creativity
-
Empowering the next generation for an AI-enabled worldby Google DeepMind Blog on November 15, 2023
Experience AI's course and resources are expanding on a global scale
-
GraphCast: AI model for faster and more accurate global weather forecastingby Google DeepMind Blog on November 14, 2023
We introduce GraphCast, a state-of-the-art AI model able to make medium-range weather forecasts with unprecedented accuracy
-
An opportunity agenda for AIby (AI) on November 14, 2023
Today we’re sharing an AI Opportunity Agenda to provide concrete policy recommendations to help AI benefit as many people as possible.
-
Taking legal action to protect users of AI and small businessesby (AI) on November 13, 2023
Today we’re taking legal action against two groups of scammers.
-
How we taught Google Translate to recognize homonymsby (AI) on November 10, 2023
How Google Translate’s neural model taught it to understand bass from bass.
-
OpenAI Data Partnershipsby OpenAI Blog on November 9, 2023
Working together to create open-source and private datasets for AI training.
-
Generative AI in Search expands to more than 120 new countries and territoriesby (AI) on November 8, 2023
Generative AI in Search, or Search Generative Experience (SGE), is expanding around the world, and adding four new languages.
-
Research Focus: Week of November 8, 2023by Alyssa Hughes (Microsoft Research) on November 8, 2023
Welcome to Research Focus, a series of blog posts that highlights notable publications, events, code/datasets, new hires and other milestones from across the research community at Microsoft. Generating both plausible and accurate full body avatar motion is essential for creating high quality immersive experiences in mixed reality scenarios. Head-mounted devices (HMDs) typically only provide a The post Research Focus: Week of November 8, 2023 appeared first on Microsoft […]
-
Boosting sustainable solutions from Swedenby (AI) on November 8, 2023
Today, we’re announcing the Swedish recipients of Google.org Impact Challenge: Tech for Social Good - receiving technical support and 3 million Euros in funding for char…
-
Toward developing faster algorithms for minimizing submodular functionsby Brenda Potts (Microsoft Research) on November 7, 2023
This research paper was presented at the 64th IEEE Symposium on Foundations of Computer Science (FOCS) 2023 (opens in new tab), a premier forum for the latest research in theoretical computer science. Submodular functions are versatile mathematical tools, finding diverse applications in real-world scenarios and guiding solutions across complex domains. From dissecting the intricate networks The post Toward developing faster algorithms for minimizing submodular functions […]
-
Introducing GPTsby OpenAI Blog on November 6, 2023
You can now create custom versions of ChatGPT that combine instructions, extra knowledge, and any combination of skills.
-
New models and developer products announced at DevDayby OpenAI Blog on November 6, 2023
GPT-4 Turbo with 128K context and lower prices, the new Assistants API, GPT-4 Turbo with Vision, DALL·E 3 API, and more.
-
A glimpse of the next generation of AlphaFoldby Google DeepMind Blog on October 31, 2023
Progress update: Our latest AlphaFold model shows significantly improved accuracy and expands coverage beyond proteins to other biological molecules, including ligands.
-
Acting on our commitment to safe and secure AIby (AI) on October 26, 2023
News on our bug bounty program specific to generative AI and how we’re supporting open source security for AI supply chains
-
Frontier risk and preparednessby OpenAI Blog on October 26, 2023
To support the safety of highly-capable AI systems, we are developing our approach to catastrophic risk preparedness, including building a Preparedness team and launching a challenge.
-
Frontier Model Forum updatesby OpenAI Blog on October 25, 2023
Together with Anthropic, Google, and Microsoft, we’re announcing the new Executive Director of the Frontier Model Forum and a new $10 million AI Safety Fund.
-
Evaluating social and ethical risks from generative AIby Google DeepMind Blog on October 19, 2023
Introducing a context-based framework for comprehensively evaluating the social and ethical risks of AI systems
-
DALL·E 3 is now available in ChatGPT Plus and Enterpriseby OpenAI Blog on October 19, 2023
We developed a safety mitigation stack to ready DALL·E 3 for wider release and are sharing updates on our provenance research.
-
How to build a secure foundation for American leadership in AIby (AI) on October 18, 2023
Google shares the report: Building a Secure Foundation for American Leadership in AI.
-
How AI can help people with motor disabilities — like my cousinby (AI) on October 17, 2023
Editor’s note: Today we’re building on our long-standing partnership with the University of Cambridge, with a multi-year research collaboration agreement and a Google gr…
-
Calling all AI cybersecurity startups — in Europe and, now, the U.S.by (AI) on October 16, 2023
Launch of the new Google for Startups Growth Academy: AI for Cybersecurity program for companies based in Europe and the U.S.
-
Our responsible approach to building guardrails for generative AIby (AI) on October 12, 2023
We’re sharing some of the ways we’re building guardrails for generative AI.
-
New ways to get inspired with generative AI in Searchby (AI) on October 12, 2023
We’re testing new ways to get started on something you need to do — like creating an image that can bring an idea to life, or a written draft when you need a starting po…
-
New ways we’re helping reduce transportation and energy emissionsby (AI) on October 10, 2023
These new product features and expansions help people, city planners and policy makers take action toward building a sustainable future.
-
Demand more from social with AI-powered adsby (AI) on October 10, 2023
Learn how Demand Gen campaigns can help you drive better results across YouTube and Google. See new case studies, videos, tips, and more.
-
HLTH 2023: Bringing AI to health responsiblyby (AI) on October 9, 2023
Google leaders take the stage at HLTH to discuss the transformative power of AI for health.
-
Android 14: More customization, control and accessibility featuresby (AI) on October 4, 2023
Android 14 is here with personal, protective and accessible features that put users first and celebrate their individuality.
-
Scaling up learning across many different robot typesby Google DeepMind Blog on October 3, 2023
Robots are great specialists, but poor generalists. Typically, you have to train a model for each task, robot, and environment. Changing a single variable often requires starting from scratch. But what if we could combine the knowledge across robotics and create a way to train a general-purpose robot?
-
DALL·E 3 system cardby OpenAI Blog on October 3, 2023
-
An update on web publisher controlsby (AI) on September 28, 2023
We’re announcing Google-Extended, a new control that web publishers can use to manage whether their sites help improve Bard and Vertex AI generative APIs, including futu…
-
New Google.org grants to introduce 300,000 students to robotics and AIby (AI) on September 27, 2023
For Google’s 25th birthday, Google.org is providing $10 million in grant funding to support robotics programs and AI education.
-
ChatGPT can now see, hear, and speakby OpenAI Blog on September 25, 2023
We are beginning to roll out new voice and image capabilities in ChatGPT. They offer a new, more intuitive type of interface by allowing you to have a voice conversation or show ChatGPT what you’re talking about.
-
GPT-4V(ision) system cardby OpenAI Blog on September 25, 2023
-
A catalogue of genetic mutations to help pinpoint the cause of diseasesby Google DeepMind Blog on September 19, 2023
New AI tool classifies the effects of 71 million ‘missense’ mutations.
-
OpenAI Red Teaming Networkby OpenAI Blog on September 19, 2023
We’re announcing an open call for the OpenAI Red Teaming Network and invite domain experts interested in improving the safety of OpenAI’s models to join our efforts.
-
Introducing OpenAI Dublinby OpenAI Blog on September 13, 2023
We’re growing our presence in Europe with an office in Dublin, Ireland.
-
Join us for OpenAI’s first developer conference on November 6 in San Franciscoby OpenAI Blog on September 6, 2023
Developer registration for in-person attendance will open in the coming weeks and developers everywhere will be able to livestream the keynote.
-
Teaching with AIby OpenAI Blog on August 31, 2023
We’re releasing a guide for teachers using ChatGPT in their classroom—including suggested prompts, an explanation of how ChatGPT works and its limitations, the efficacy of AI detectors, and bias.
-
Identifying AI-generated images with SynthIDby Google DeepMind Blog on August 29, 2023
New tool helps watermark and identify synthetic images created by Imagen
-
Introducing ChatGPT Enterpriseby OpenAI Blog on August 28, 2023
Get enterprise-grade security & privacy and the most powerful version of ChatGPT yet.
-
OpenAI partners with Scale to provide support for enterprises fine-tuning modelsby OpenAI Blog on August 24, 2023
OpenAI’s customers can leverage Scale’s AI expertise to customize our most advanced models.
-
GPT-3.5 Turbo fine-tuning and API updatesby OpenAI Blog on August 22, 2023
Developers can now bring their own data to customize GPT-3.5 Turbo for their use cases.
-
OpenAI acquires Global Illuminationby OpenAI Blog on August 16, 2023
The entire team has joined OpenAI.
-
Using GPT-4 for content moderationby OpenAI Blog on August 15, 2023
We use GPT-4 for content policy development and content moderation decisions, enabling more consistent labeling, a faster feedback loop for policy refinement, and less involvement from human moderators.
-
RT-2: New model translates vision and language into actionby Google DeepMind Blog on July 28, 2023
Robotic Transformer 2 (RT-2) is a novel vision-language-action (VLA) model that learns from both web and robotics data, and translates this knowledge into generalised instructions for robotic control.
-
Frontier Model Forumby OpenAI Blog on July 26, 2023
We’re forming a new industry body to promote the safe and responsible development of frontier AI systems: advancing AI safety research, identifying best practices and standards, and facilitating information sharing among policymakers and industry.
-
Moving AI governance forwardby OpenAI Blog on July 21, 2023
OpenAI and other leading labs reinforce AI safety, security and trustworthiness through voluntary commitments.
-
Using AI to fight climate changeby Google DeepMind Blog on July 21, 2023
AI is a powerful technology that will transform our future, so how can we best apply it to help combat climate change and find sustainable solutions?
-
Custom instructions for ChatGPTby OpenAI Blog on July 20, 2023
We’re rolling out custom instructions to give you more control over how ChatGPT responds. Set your preferences, and ChatGPT will keep them in mind for all future conversations.
-
Google DeepMind’s latest research at ICML 2023by Google DeepMind Blog on July 20, 2023
Exploring AI safety, adaptability, and efficiency for the real world
-
Developing reliable AI tools for healthcareby Google DeepMind Blog on July 17, 2023
We’ve published our joint paper with Google Research in Nature Medicine, which proposes CoDoC (Complementarity-driven Deferral-to-Clinical Workflow), an AI system that learns when to rely on predictive AI tools or defer to a clinician for the most accurate interpretation of medical images.
-
Exploring institutions for global AI governanceby Google DeepMind Blog on July 11, 2023
New white paper investigates models and functions of international institutions that could help manage opportunities and mitigate risks of advanced AI.
-
Frontier AI regulation: Managing emerging risks to public safetyby OpenAI Blog on July 6, 2023
-
RoboCat: A self-improving robotic agentby Google DeepMind Blog on June 20, 2023
Robots are quickly becoming part of our everyday lives, but they’re often only programmed to perform specific tasks well. While harnessing recent advances in AI could lead to robots that could help in many more ways, progress in building general-purpose robots is slower in part because of the time needed to collect real-world training data. Our latest paper introduces a self-improving AI agent for robotics, RoboCat, that learns to perform a variety of tasks across […]
-
AlphaDev discovers faster sorting algorithmsby Google DeepMind Blog on June 7, 2023
New algorithms will transform the foundations of computing
-
Improving mathematical reasoning with process supervisionby OpenAI Blog on May 31, 2023
We've trained a model to achieve a new state-of-the-art in mathematical problem solving by rewarding each correct step of reasoning (“process supervision”) instead of simply rewarding the correct final answer (“outcome supervision”). In addition to boosting performance relative to outcome supervision, process supervision also has an important alignment benefit: it directly trains the model to produce a chain-of-thought that is endorsed by humans.
-
An early warning system for novel AI risksby Google DeepMind Blog on May 25, 2023
New research proposes a framework for evaluating general-purpose models against novel threats
-
How generational differences affect consumer attitudes towards adsby Meta Research on May 17, 2023
Our research study, in collaboration with CrowdDNA, aims to understand people's relationship with social media ads across different social media platforms.
-
Language models can explain neurons in language modelsby OpenAI Blog on May 9, 2023
We use GPT-4 to automatically write explanations for the behavior of neurons in large language models and to score those explanations. We release a dataset of these (imperfect) explanations and scores for every neuron in GPT-2.
-
DeepMind’s latest research at ICLR 2023by Google DeepMind Blog on April 27, 2023
Next week marks the start of the 11th International Conference on Learning Representations (ICLR), taking place 1-5 May in Kigali, Rwanda. This will be the first major artificial intelligence (AI) conference to be hosted in Africa and the first in-person event since the start of the pandemic. Researchers from around the world will gather to share their cutting-edge work in deep learning spanning the fields of AI, statistics and data science, and applications including […]
-
How can we build human values into AI?by Google DeepMind Blog on April 24, 2023
Drawing from philosophy to identify fair principles for ethical AI...
-
Announcing Google DeepMindby Google DeepMind Blog on April 20, 2023
DeepMind and the Brain team from Google Research will join forces to accelerate progress towards a world in which AI helps solve the biggest challenges facing humanity.
-
Every tree countsby Meta Research on April 17, 2023
Meta set a goal to reach net zero emissions by 2030. We are developing technology to mitigate our carbon footprint and making these openly available.
-
How a non-traditional background led to cutting-edge XR techby Meta Research on April 14, 2023
-
A new, unique AI dataset for animating amateur drawingsby Meta Research on April 13, 2023
-
How the metaverse can transform educationby Meta Research on April 12, 2023
-
Build faster with Buck2: Our open source build systemby Meta Research on April 6, 2023
-
Announcing the 2023 Meta Research PhD Fellowship award winnersby Meta Research on April 5, 2023
...
-
Announcing the winners of the 2022 Foundational Integrity Research request for proposalsby Meta Research on March 27, 2023
In September, Meta launched the Foundational Integrity Research request for proposals. Today, we announce the winners of this award.
-
Two meta sustainability grant and scholarship recipients share impactby Meta Research on March 24, 2023
-
GPT-4by OpenAI Blog on March 14, 2023
We’ve created GPT-4, the latest milestone in OpenAI’s effort in scaling up deep learning. GPT-4 is a large multimodal model (accepting image and text inputs, emitting text outputs) that, while less capable than humans in many real-world scenarios, exhibits human-level performance on various professional and academic benchmarks.
-
Forecasting potential misuses of language models for disinformation campaigns and how to reduce riskby OpenAI Blog on January 11, 2023
OpenAI researchers collaborated with Georgetown University’s Center for Security and Emerging Technology and the Stanford Internet Observatory to investigate how large language models might be misused for disinformation purposes. The collaboration included an October 2021 workshop bringing together 30 disinformation researchers, machine learning experts, and policy analysts, and culminated in a co-authored report building on more than a year of research. This report […]
-
Point-E: A system for generating 3D point clouds from complex promptsby OpenAI Blog on December 16, 2022
-
Competitive programming with AlphaCodeby Google DeepMind Blog on December 8, 2022
Solving novel problems and setting a new milestone in competitive programming.
-
AI for the board game Diplomacyby Google DeepMind Blog on December 6, 2022
Successful communication and cooperation have been crucial for helping societies advance throughout history. The closed environments of board games can serve as a sandbox for modelling and investigating interaction and communication – and we can learn a lot from playing them. In our recent paper, published today in Nature Communications, we show how artificial agents can use communication to better cooperate in the board game Diplomacy, a vibrant domain in artificial […]
-
Mastering Stratego, the classic game of imperfect informationby Google DeepMind Blog on December 1, 2022
Game-playing artificial intelligence (AI) systems have advanced to a new frontier.
-
DeepMind’s latest research at NeurIPS 2022by Google DeepMind Blog on November 25, 2022
NeurIPS is the world’s largest conference in artificial intelligence (AI) and machine learning (ML), and we’re proud to support the event as Diamond sponsors, helping foster the exchange of research advances in the AI and ML community. Teams from across DeepMind are presenting 47 papers, including 35 external collaborations in virtual panels and poster sessions.
-
Building interactive agents in video game worldsby Google DeepMind Blog on November 23, 2022
Most artificial intelligence (AI) researchers now believe that writing computer code which can capture the nuances of situated interactions is impossible. Alternatively, modern machine learning (ML) researchers have focused on learning about these types of interactions from data. To explore these learning-based approaches and quickly build agents that can make sense of human instructions and safely perform actions in open-ended conditions, we created a research framework […]
-
Benchmarking the next generation of never-ending learnersby Google DeepMind Blog on November 22, 2022
Learning how to build upon knowledge by tapping 30 years of computer vision research
-
Best practices for data enrichmentby Google DeepMind Blog on November 16, 2022
Building a responsible approach to data collection with the Partnership on AI...
-
The pursuit of AI education—past, present, and futureby Google DeepMind Blog on November 8, 2022
Meet Sylvia Christie, our education partnerships manager who’s played a leading role in expanding our scholarship programme, which is marking its five-year anniversary.
-
Scaling laws for reward model overoptimizationby OpenAI Blog on October 19, 2022
-
Stopping malaria in its tracksby Google DeepMind Blog on October 13, 2022
Developing a vaccine that could save hundreds of thousands of lives
-
Measuring perception in AI modelsby Google DeepMind Blog on October 12, 2022
Perception – the process of experiencing the world through senses – is a significant part of intelligence. And building agents with human-level perceptual understanding of the world is a central but challenging task, which is becoming increasingly important in robotics, self-driving cars, personal assistants, medical imaging, and more. So today, we’re introducing the Perception Test, a multimodal benchmark using real-world videos to help evaluate the perception […]
-
How undesired goals can arise with correct rewardsby Google DeepMind Blog on October 7, 2022
As we build increasingly advanced artificial intelligence (AI) systems, we want to make sure they don’t pursue undesired goals. Such behaviour in an AI agent is often the result of specification gaming – exploiting a poor choice of what they are rewarded for. In our latest paper, we explore a more subtle mechanism by which AI systems may unintentionally learn to pursue undesired goals: goal misgeneralisation (GMG). GMG occurs when a system's capabilities generalise […]
-
Discovering novel algorithms with AlphaTensorby Google DeepMind Blog on October 5, 2022
In our paper, published today in Nature, we introduce AlphaTensor, the first artificial intelligence (AI) system for discovering novel, efficient, and provably correct algorithms for fundamental tasks such as matrix multiplication. This sheds light on a 50-year-old open question in mathematics about finding the fastest way to multiply two matrices. This paper is a stepping stone in DeepMind’s mission to advance science and unlock the most fundamental problems using AI. Our […]
-
Fighting osteoporosis before it startsby Google DeepMind Blog on September 27, 2022
Detecting signs of disease before bones start to break
-
Understanding the faulty proteins linked to cancer and autismby Google DeepMind Blog on September 26, 2022
Helping uncover how protein mutations cause diseases and disorders
-
Supporting the next generation of AI leadersby Google DeepMind Blog on September 26, 2022
We’re partnering with six education charities and social enterprises in the United Kingdom (UK) to co-create a bespoke education programme to help tackle the gaps in STEM education and boost existing programmes.
-
Solving the mystery of how an ancient bird went extinctby Google DeepMind Blog on September 22, 2022
Creating a tool to study extinct species from 50,000 years ago
-
Building safer dialogue agentsby Google DeepMind Blog on September 22, 2022
In our latest paper, we introduce Sparrow – a dialogue agent that’s useful and reduces the risk of unsafe and inappropriate answers. Our agent is designed to talk with a user, answer questions, and search the internet using Google when it’s helpful to look up evidence to inform its responses.
-
Targeting early-onset Parkinson’s with AIby Google DeepMind Blog on September 21, 2022
Predictions that pave the way to new treatments
-
Introducing Whisperby OpenAI Blog on September 21, 2022
We’ve trained and are open-sourcing a neural net called Whisper that approaches human level robustness and accuracy on English speech recognition.
-
How our principles helped define AlphaFold’s releaseby Google DeepMind Blog on September 14, 2022
Our Operating Principles have come to define both our commitment to prioritising widespread benefit, as well as the areas of research and applications we refuse to pursue. These principles have been at the heart of our decision making since DeepMind was founded, and continue to be refined as the AI landscape changes and grows. They are designed for our role as a research-driven science company and consistent with Google’s AI principles.
-
Maximising the impact of our breakthroughsby Google DeepMind Blog on September 9, 2022
Colin, CBO at DeepMind, discusses collaborations with Alphabet and how we integrate ethics, accountability, and safety into everything we do.
-
My journey from DeepMind intern to mentorby Google DeepMind Blog on September 8, 2022
Former intern turned intern manager, Richard Everett, describes his journey to DeepMind, sharing tips and advice for aspiring DeepMinders. The 2023 internship applications will open on the 16th September, please visit https://dpmd.ai/internshipsatdeepmind for more information.
-
In conversation with AI: building better language modelsby Google DeepMind Blog on September 6, 2022
Our new paper, In conversation with AI: aligning language models with human values, explores a different approach, asking what successful communication between humans and an artificial conversational agent might look like and what values should guide conversation in these contexts.
-
From motor control to embodied intelligenceby Google DeepMind Blog on August 31, 2022
Using human and animal motions to teach robots to dribble a ball, and simulated humanoid characters to carry boxes and play football
-
Advancing conservation with AI-based facial recognition of turtlesby Google DeepMind Blog on August 25, 2022
We came across Zindi – a dedicated partner with complementary goals – who are the largest community of African data scientists and host competitions that focus on solving Africa’s most pressing problems. Our Science team’s Diversity, Equity, and Inclusion (DE&I) team worked with Zindi to identify a scientific challenge that could help advance conservation efforts and grow involvement in AI. Inspired by Zindi’s bounding box turtle challenge, we landed on a […]
-
Discovering when an agent is present in a systemby Google DeepMind Blog on August 18, 2022
We want to build safe, aligned artificial general intelligence (AGI) systems that pursue the intended goals of its designers. Causal influence diagrams (CIDs) are a way to model decision-making situations that allow us to reason about agent incentives. By relating training setups to the incentives that shape agent behaviour, CIDs help illuminate potential risks before training an agent and can inspire better agent designs. But how do we know when a CID is an accurate model […]
-
Realising scientists are the real superheroesby Google DeepMind Blog on August 11, 2022
Meet Edgar Duéñez-Guzmán, a research engineer on our Multi-Agent Research team who’s drawing on knowledge of game theory, computer science, and social evolution to get AI agents working better together.
-
The race to cure a billion people from a deadly parasitic diseaseby Google DeepMind Blog on July 28, 2022
Accelerating the search for life saving leishmaniasis treatments
-
Tracing the evolution of proteins back to the origin of lifeby Google DeepMind Blog on July 28, 2022
Looking into a protein’s past to unlock the mysteries of life itself
-
How the honeybee could help protect species around the worldby Google DeepMind Blog on July 28, 2022
New insights into immunity to help protect the world’s flora
-
AlphaFold transforms biology for millions around the worldby Google DeepMind Blog on July 28, 2022
Big data that leads to discoveries that benefit everyone
-
Advancing discovery of better drugs and medicineby Google DeepMind Blog on July 28, 2022
Researchers are designing more effective drugs than ever before
-
Creating plastic-eating enzymes that could save us from pollutionby Google DeepMind Blog on July 28, 2022
Helping plastics become 100% recyclable
-
AlphaFold unlocks one of the greatest puzzles in biologyby Google DeepMind Blog on July 28, 2022
Piecing together one of the largest molecular structures in human cells
-
Accelerating the race against antibiotic resistanceby Google DeepMind Blog on July 28, 2022
Unlocking a decade of data in minutes to help beat antibiotic resistance
-
Efficient training of language models to fill in the middleby OpenAI Blog on July 28, 2022
-
AlphaFold reveals the structure of the protein universeby Google DeepMind Blog on July 28, 2022
Today, in partnership with EMBL’s European Bioinformatics Institute (EMBL-EBI), we’re now releasing predicted structures for nearly all catalogued proteins known to science, which will expand the AlphaFold DB by over 200x - from nearly 1 million structures to over 200 million structures - with the potential to dramatically increase our understanding of biology.
-
A hazard analysis framework for code synthesis large language modelsby OpenAI Blog on July 25, 2022
-
Putting the power of AlphaFold into the world’s handsby Google DeepMind Blog on July 22, 2022
When we announced AlphaFold 2 last December, it was hailed as a solution to the 50-year old protein folding problem. Last week, we published the scientific paper and source code explaining how we created this highly innovative system, and today we’re sharing high-quality predictions for the shape of every single protein in the human body, as well as for the proteins of 20 additional organisms that scientists rely on for their research.
-
The virtuous cycle of AI researchby Google DeepMind Blog on July 19, 2022
We recently caught up with Petar Veličković, a research scientist at DeepMind. Along with his co-authors, Petar is presenting his paper The CLRS Algorithmic Reasoning Benchmark at ICML 2022 in Baltimore, Maryland, USA.
-
Perceiver AR: general-purpose, long-context autoregressive generationby Google DeepMind Blog on July 16, 2022
We develop Perceiver AR, an autoregressive, modality-agnostic architecture which uses cross-attention to map long-range inputs to a small number of latents while also maintaining end-to-end causal masking. Perceiver AR can directly attend to over a hundred thousand tokens, enabling practical long-context density estimation without the need for hand-crafted sparsity patterns or memory mechanisms.
-
DeepMind’s latest research at ICML 2022by Google DeepMind Blog on July 15, 2022
Starting this weekend, the thirty-ninth International Conference on Machine Learning (ICML 2022) is meeting from 17-23 July, 2022 at the Baltimore Convention Center in Maryland, USA, and will be running as a hybrid event. Researchers working across artificial intelligence, data science, machine vision, computational biology, speech recognition, and more are presenting and publishing their cutting-edge work in machine learning.
-
Intuitive physics learning in a deep-learning model inspired by developmental psychologyby Google DeepMind Blog on July 11, 2022
Despite significant effort, current AI systems pale in their understanding of intuitive physics, in comparison to even very young children. In the present work, we address this AI problem, specifically by drawing on the field of developmental psychology.
-
Human-centred mechanism design with Democratic AIby Google DeepMind Blog on July 4, 2022
In our recent paper, published in Nature Human Behaviour, we provide a proof-of-concept demonstration that deep reinforcement learning (RL) can be used to find economic policies that people will vote for by majority in a simple game. The paper thus addresses a key challenge in AI research - how to train AI systems that align with human values.
-
DALL·E 2 pre-training mitigationsby OpenAI Blog on June 28, 2022
In order to share the magic of DALL·E 2 with a broad audience, we needed to reduce the risks associated with powerful image generation models. To this end, we put various guardrails in place to prevent generated images from violating our content policy.
-
Learning to play Minecraft with Video PreTrainingby OpenAI Blog on June 23, 2022
We trained a neural network to play Minecraft by Video PreTraining (VPT) on a massive unlabeled video dataset of human Minecraft play, while using only a small amount of labeled contractor data. With fine-tuning, our model can learn to craft diamond tools, a task that usually takes proficient humans over 20 minutes (24,000 actions). Our model uses the native human interface of keypresses and mouse movements, making it quite general, and represents a step towards general […]
-
Leading a movement to strengthen machine learning in Africaby Google DeepMind Blog on June 23, 2022
Avishkar Bhoopchand, a research engineer on the Game Theory and Multi-agent team, shares his journey to DeepMind and how he’s working to raise the profile of deep learning across Africa.
-
BYOL-Explore: Exploration with Bootstrapped Predictionby Google DeepMind Blog on June 20, 2022
We present BYOL-Explore, a conceptually simple yet general approach for curiosity-driven exploration in visually-complex environments. BYOL-Explore learns a world representation, the world dynamics, and an exploration policy all-together by optimizing a single prediction loss in the latent space with no additional auxiliary objective. We show that BYOL-Explore is effective in DM-HARD-8, a challenging partially-observable continuous-action hard-exploration benchmark with […]
-
Evolution through large modelsby OpenAI Blog on June 17, 2022
-
Unlocking High-Accuracy Differentially Private Image Classification through Scaleby Google DeepMind Blog on June 17, 2022
According to empirical evidence from prior works, utility degradation in DP-SGD becomes more severe on larger neural network models – including the ones regularly used to achieve the best performance on challenging image classification benchmarks. Our work investigates this phenomenon and proposes a series of simple modifications to both the training procedure and model architecture, yielding a significant improvement on the accuracy of DP training on standard image […]
-
Bridging DeepMind research with Alphabet productsby Google DeepMind Blog on June 15, 2022
Today we caught up with Gemma Jennings, a product manager on the Applied team, who led a session on vision language models at the AI Summit, one of the world’s largest AI events for business.
-
AI-written critiques help humans notice flawsby OpenAI Blog on June 13, 2022
We trained “critique-writing” models to describe flaws in summaries. Human evaluators find flaws in summaries much more often when shown our model’s critiques. Larger models are better at self-critiquing, with scale improving critique-writing more than summary-writing. This shows promise for using AI systems to assist human supervision of AI systems on difficult tasks.
-
Techniques for training large neural networksby OpenAI Blog on June 9, 2022
Large neural networks are at the core of many recent advances in AI, but training them is a difficult engineering and research challenge which requires orchestrating a cluster of GPUs to perform a single synchronized calculation.
-
Advocating for the LGBTQ+ community in AI researchby Google DeepMind Blog on June 1, 2022
Research scientist, Kevin McKee, tells how his early love of science fiction and social psychology inspired his career, and how he’s helping advance research in ‘queer fairness’, support human-AI collaboration, and study the effects of AI on the LGBTQ+ community.
-
Teaching models to express their uncertainty in wordsby OpenAI Blog on May 28, 2022
-
Evaluating Multimodal Interactive Agentsby Google DeepMind Blog on May 27, 2022
In this paper, we assess the merits of these existing evaluation metrics and present a novel approach to evaluation called the Standardised Test Suite (STS). The STS uses behavioural scenarios mined from real human interaction data.
-
Dynamic language understanding: adaptation to new knowledge in parametric and semi-parametric modelsby Google DeepMind Blog on May 26, 2022
To study how semi-parametric QA models and their underlying parametric language models (LMs) adapt to evolving knowledge, we construct a new large-scale dataset, StreamingQA, with human written and generated questions asked on a given date, to be answered from 14 years of time-stamped news articles. We evaluate our models quarterly as they read new articles not seen in pre-training. We show that parametric models can be updated without full retraining, while avoiding […]
-
Kyrgyzstan to King’s Cross: the star baker cooking up codeby Google DeepMind Blog on May 26, 2022
My day can vary, it really depends on which phase of the project I'm on. Let’s say we want to add a feature to our product – my tasks could range from designing solutions and working with the team to find the best one, to deploying new features into production and doing maintenance. Along the way, I’ll communicate changes to our stakeholders, write docs, code and test solutions, build analytics dashboards, clean-up old code, and fix bugs.
-
Building a culture of pioneering responsiblyby Google DeepMind Blog on May 24, 2022
When I joined DeepMind as COO, I did so in large part because I could tell that the founders and team had the same focus on positive social impact. In fact, at DeepMind, we now champion a term that perfectly captures my own values and hopes for integrating technology into people’s daily lives: pioneering responsibly. I believe pioneering responsibly should be a priority for anyone working in tech. But I also recognise that it’s especially important when it comes to […]
-
Open-sourcing MuJoCoby Google DeepMind Blog on May 23, 2022
In October 2021, we announced that we acquired the MuJoCo physics simulator, and made it freely available for everyone to support research everywhere. We also committed to developing and maintaining MuJoCo as a free, open-source, community-driven project with best-in-class capabilities. Today, we’re thrilled to report that open sourcing is complete and the entire codebase is on GitHub! Here, we explain why MuJoCo is a great platform for open-source collaboration and share […]
-
From LEGO competitions to DeepMind's robotics labby Google DeepMind Blog on May 19, 2022
If you want to be at DeepMind, go for it. Apply, interview, and just try. You might not get it the first time but that doesn’t mean you can’t try again. I never thought DeepMind would accept me, and when they did, I thought it was a mistake. Everyone doubts themselves – I’ve never felt like the smartest person in the room. I’ve often felt the opposite. But I’ve learned that, despite those feelings, I do belong and I do deserve to work at a place like this. And […]
-
Emergent Bartering Behaviour in Multi-Agent Reinforcement Learningby Google DeepMind Blog on May 16, 2022
In our recent paper, we explore how populations of deep reinforcement learning (deep RL) agents can learn microeconomic behaviours, such as production, consumption, and trading of goods. We find that artificial agents learn to make economically rational decisions about production, consumption, and prices, and react appropriately to supply and demand changes.
-
A Generalist Agentby Google DeepMind Blog on May 12, 2022
Inspired by progress in large-scale language modelling, we apply a similar approach towards building a single generalist agent beyond the realm of text outputs. The agent, which we refer to as Gato, works as a multi-modal, multi-task, multi-embodiment generalist policy. The same network with the same weights can play Atari, caption images, chat, stack blocks with a real robot arm and much more, deciding based on its context whether to output text, joint torques, button […]
-
Active offline policy selectionby Google DeepMind Blog on May 6, 2022
To make RL more applicable to real-world applications like robotics, we propose using an intelligent evaluation procedure to select the policy for deployment, called active offline policy selection (A-OPS). In A-OPS, we make use of the prerecorded dataset and allow limited interactions with the real environment to boost the selection quality.
-
When a passion for bass and brass help build better toolsby Google DeepMind Blog on April 28, 2022
We caught up with Kevin Millikin, a software engineer on the DevTools team. He’s in Salt Lake City this week to present at PyCon US, the largest annual gathering for those using and developing the open-source Python programming language.
-
Tackling multiple tasks with a single visual language modelby Google DeepMind Blog on April 28, 2022
We introduce Flamingo, a single visual language model (VLM) that sets a new state of the art in few-shot learning on a wide range of open-ended multimodal tasks.
-
DeepMind’s latest research at ICLR 2022by Google DeepMind Blog on April 25, 2022
Beyond supporting the event as sponsors and regular workshop organisers, our research teams are presenting 29 papers, including 10 collaborations this year. Here’s a brief glimpse into our upcoming oral, spotlight, and poster presentations.
-
An empirical analysis of compute-optimal large language model trainingby Google DeepMind Blog on April 12, 2022
We ask the question: “What is the optimal model size and number of training tokens for a given compute budget?” To answer this question, we train models of various sizes and with various numbers of tokens, and estimate this trade-off empirically. Our main finding is that the current large language models are far too large for their compute budget and are not being trained on enough data.
-
GopherCite: Teaching language models to support answers with verified quotesby Google DeepMind Blog on March 16, 2022
Language models like Gopher can “hallucinate” facts that appear plausible but are actually fake. Those who are familiar with this problem know to do their own fact-checking, rather than trusting what language models say. Those who are not, may end up believing something that isn’t true. This paper describes GopherCite, a model which aims to address the problem of language model hallucination. GopherCite attempts to back up all of its factual claims with evidence from […]
-
Predicting the past with Ithacaby Google DeepMind Blog on March 9, 2022
The birth of human writing marked the dawn of History and is crucial to our understanding of past civilisations and the world we live in today. For example, more than 2,500 years ago, the Greeks began writing on stone, pottery, and metal to document everything from leases and laws to calendars and oracles, giving a detailed insight into the Mediterranean region. Unfortunately, it’s an incomplete record. Many of the surviving inscriptions have been damaged over the […]
-
Learning Robust Real-Time Cultural Transmission without Human Databy Google DeepMind Blog on March 3, 2022
In this work, we use deep reinforcement learning to generate artificial agents capable of test-time cultural transmission. Once trained, our agents can infer and recall navigational knowledge demonstrated by experts. This knowledge transfer happens in real time and generalises across a vast space of previously unseen tasks.
-
Probing Image-Language Transformers for Verb Understandingby Google DeepMind Blog on February 23, 2022
Multimodal Image-Language transformers have achieved impressive results on a variety of tasks that rely on fine-tuning (e.g., visual question answering and image retrieval). We are interested in shedding light on the quality of their pretrained representations--in particular, if these models can distinguish verbs or they only use the nouns in a given sentence. To do so, we collect a dataset of image-sentence pairs consisting of 447 verbs that are either visual or commonly […]
-
Accelerating fusion science through learned plasma controlby Google DeepMind Blog on February 16, 2022
Successfully controlling the nuclear fusion plasma in a tokamak with deep reinforcement learning
-
MuZero’s first step from research into the real worldby Google DeepMind Blog on February 11, 2022
Collaborating with YouTube to optimise video compression in the open source VP9 codec.
-
Red Teaming Language Models with Language Modelsby Google DeepMind Blog on February 7, 2022
In our recent paper, we show that it is possible to automatically find inputs that elicit harmful text from language models by generating inputs using language models themselves. Our approach provides one tool for finding harmful model behaviours before users are impacted, though we emphasize that it should be viewed as one component alongside many other techniques that will be needed to find harms and mitigate them once found.
-
DeepMind: The Podcast returns for Season 2by Google DeepMind Blog on January 25, 2022
We believe artificial intelligence (AI) is one of the most significant technologies of our age and we want to help people understand its potential and how it’s being created.
-
Spurious normativity enhances learning of compliance and enforcement behavior in artificial agentsby Google DeepMind Blog on January 18, 2022
In our recent paper we explore how multi-agent deep reinforcement learning can serve as a model of complex social interactions, like the formation of social norms. This new class of models could provide a path to create richer, more detailed simulations of the world.
-
AlphaFold: Using AI for scientific discoveryby Google DeepMind Blog on January 15, 2022
We’re excited to share DeepMind’s first significant milestone in demonstrating how artificial intelligence research can drive and accelerate new scientific discoveries.
-
Simulating matter on the quantum scale with AIby Google DeepMind Blog on December 9, 2021
Solving some of the major challenges of the 21st Century, such as producing clean electricity or developing high temperature superconductors, will require us to design new materials with specific properties. To do this on a computer requires the simulation of electrons, the subatomic particles that govern how atoms bond to form molecules and are also responsible for the flow of electricity in solids.
-
Language modelling at scale: Gopher, ethical considerations, and retrievalby Google DeepMind Blog on December 8, 2021
Language, and its role in demonstrating and facilitating comprehension - or intelligence - is a fundamental part of being human. It gives people the ability to communicate thoughts and concepts, express ideas, create memories, and build mutual understanding. These are foundational parts of social intelligence. It’s why our teams at DeepMind study aspects of language processing and communication, both in artificial agents and in humans.
-
Improving language models by retrieving from trillions of tokensby Google DeepMind Blog on December 8, 2021
We explore an alternate path for improving language models: we augment transformers with retrieval over a database of text passages including web pages, books, news and code. We call our method RETRO, for “Retrieval Enhanced TRansfOrmers”.
-
Creating Interactive Agents with Imitation Learningby Google DeepMind Blog on December 8, 2021
We show that imitation learning of human-human interactions in a simulated world, in conjunction with self-supervised learning, is sufficient to produce a multimodal interactive agent, which we call MIA, that successfully interacts with non-adversarial humans 75% of the time. We further identify architectural and algorithmic techniques that improve performance, such as hierarchical action selection.
-
Exploring the beauty of pure mathematics in novel waysby Google DeepMind Blog on December 1, 2021
More than a century ago, Srinivasa Ramanujan shocked the mathematical world with his extraordinary ability to see remarkable patterns in numbers that no one else could see. The self-taught mathematician from India described his insights as deeply intuitive and spiritual, and patterns often came to him in vivid dreams.
-
On the Expressivity of Markov Rewardby Google DeepMind Blog on December 1, 2021
Our main results prove that while reward can express many tasks, there exist instances of each task type that no Markov reward function can capture. We then provide a set of polynomial-time algorithms that construct a reward function which allows an agent to optimize tasks of each of these three types, and correctly determine when no such reward function exists.
-
Unsupervised deep learning identifies semantic disentanglement in single inferotemporal face patch neuronsby Google DeepMind Blog on November 9, 2021
Our brain has an amazing ability to process visual information. We can take one glance at a complex scene, and within milliseconds be able to parse it into objects and their attributes, like colour or size, and use this information to describe the scene in simple language. Underlying this seemingly effortless ability is a complex computation performed by our visual cortex, which involves taking millions of neural impulses transmitted from the retina and transforming them […]
-
Real-world challenges for AGIby Google DeepMind Blog on November 2, 2021
When people picture a world with artificial general intelligence (AGI), robots are more likely to come to mind than enabling solutions to society’s most intractable problems. But I believe the latter is much closer to the truth. AI is already enabling huge leaps in tackling fundamental challenges: from solving protein folding to predicting accurate weather patterns, scientists are increasingly using AI to deduce the rules and principles that underpin highly complex […]
-
Opening up a physics simulator for roboticsby Google DeepMind Blog on October 18, 2021
When you walk, your feet make contact with the ground. When you write, your fingers make contact with the pen. Physical contacts are what makes interaction with the world possible. Yet, for such a common occurrence, contact is a surprisingly complex phenomenon. Taking place at microscopic scales at the interface of two bodies, contacts can be soft or stiff, bouncy or spongy, slippery or sticky. It’s no wonder our fingertips have four different types of touch-sensors. This […]
-
Stacking our way to more general robotsby Google DeepMind Blog on October 11, 2021
Picking up a stick and balancing it atop a log or stacking a pebble on a stone may seem like simple — and quite similar — actions for a person. However, most robots struggle with handling more than one such task at a time. Manipulating a stick requires a different set of behaviours than stacking stones, never mind piling various dishes on top of one another or assembling furniture. Before we can teach robots how to perform these kinds of tasks, they first need to learn […]
-
Predicting gene expression with AIby Google DeepMind Blog on October 4, 2021
When the Human Genome Project succeeded in mapping the DNA sequence of the human genome, the international research community were excited by the opportunity to better understand the genetic instructions that influence human health and development. DNA carries the genetic information that determines everything from eye colour to susceptibility to certain diseases and disorders. The roughly 20,000 sections of DNA in the human body known as genes contain instructions about the […]