OpenAI Five is the first AI to beat the world champions in an esports game, having won two back-to-back games versus the world champion Dota 2 team, OG [https://twitter.com/OGesports], at Finals…

We’ll be holding our final live event for OpenAI Five at 11:30a PT on April 13th.

Energy based models (EBMs) are appealing due to their generality and simplicity in likelihood modeling, but have been traditionally difficult to train. We present techniques to scale MCMC based EBM…

We've made progress towards stable and scalable training of energy-based models [http://yann.lecun.com/exdb/publis/pdf/lecun-06.pdf] (EBMs) resulting in better sample quality and generalization…

Our class of eight Scholars (out of 550 applicants) brings together collective expertise in literature, philosophy, cell biology, statistics, economics, quantum physics, and business innovation. Our…

We’ve created OpenAI LP, a new “capped-profit” company that allows us to rapidly increase our investments in compute and talent while including checks and balances to actualize our mission.

lucid - A collection of infrastructure and tools for research in neural network interpretability.

By using feature inversion to visualize millions of activations from an image classification network, we create an explorable activation atlas of features the network has learned and what concepts it…

We’ve created activation atlases (in collaboration with Google researchers), a new technique for visualizing what interactions between neurons can represent.

3D Client for https://github.com/openai/neural-mmo - jsuarez5341/neural-mmo-client

Neural MMO - A Massively Multiagent Game Environment - openai/neural-mmo

We’re releasing our Neural MMO - a massively multiagent game environment for reinforcement learning agents.

For educational materials related to the spinning up workshops. - openai/spinningup-workshop

On February 2nd, we hosted our first Spinning Up Workshop as part of our new OpenAI Education Initiative. This workshop built off our Spinning Up in Deep RL resource package and took a deeper dive…

We've written a paper arguing that long-term AI safety research needs social scientists to ensure AI alignment algorithms succeed when actual humans are involved.

We’ve trained a large-scale unsupervised language model which generates coherent paragraphs of text, achieves state-of-the-art performance on many language modeling benchmarks, and performs rudimentary reading comprehension, machine translation, question answering, and summarization — all without task-specific training. View codeRead paperRead more Our model, called

09:00AM - 10:30PM: Opening & Intro to RL, Part 1 10:30AM - 10:45PM: Untelevised Coffee Break 10:45AM - 11:30PM: Intro to RL, Part 2 11:30AM - 12:00PM: Safety

Our first cohort of OpenAI fellows have completed their final projects, with each Fellow going from a machine learning beginner to core OpenAI contributor.

We've discovered that the gradient noise scale, a simple statistical metric, predicts the parallelizability of neural network training.

We’re releasing a new training environment, CoinRun, that precisely quantifies an agent’s ability to transfer its experience to novel test environments.

We’re releasing Spinning Up in Deep RL, an educational resource designed to let anyone learn to become a skilled practitioner in deep reinforcement learning. Spinning Up consists of crystal-clear…

We’ve built an energy-based model that can quickly recognize, generate, and transfer simple concepts (such as near, above, between, closest, and furthest) represented by sequences or sets of 2D…

(Submitted on 5 Nov 2018) Abstract: We propose a plan online and learn offline (POLO) framework for the setting where an agent, with an internal model, needs to continually act and learn in the world.…