Large Language Models

We’ve trained a human-like robot hand to manipulate physical objects with unprecedented dexterity.

Our first class of OpenAI Scholars is underway, and you can now follow along as this group of experienced software developers becomes machine learning practitio

A simple and surprisingly effective family of conditioning mechanisms.

We introduce Glow, a reversible generative model which uses invertible 1x1 convolutions. It extends previous work on reversible generative models and simplifies

We’ve trained an agent to achieve a high score of 74,500 on Montezuma’s Revenge from a single human demonstration, better than any previously published result.

Our team of five neural networks, OpenAI Five, has started to defeat amateur human teams at Dota 2.

The first run of our Retro Contest—exploring the development of algorithms that can generalize from previous experience—is now complete.

We’re now accepting applications for the next cohort of OpenAI Fellows, a program which offers a compensated 6-month apprenticeship in AI research at OpenAI.

We’re releasing the full version of Gym Retro, a platform for reinforcement learning research on games. This brings our publicly-released game count from around

We’re releasing an analysis showing that since 2012, the amount of compute used in the largest AI training runs has been increasing exponentially with a 3.4-mon

The full implementation is available in lilianweng/deep-reinforcement-learning-gym In the previous two posts, I have introduced the algorithms of many deep rein

We’re proposing an AI safety technique which trains agents to debate topics with one another, using a human to judge who wins.

We’re releasing an experimental metalearning approach called Evolved Policy Gradients, a method that evolves the loss function of learning agents, which can ena

[Updated on 2018-06-30: add two new policy gradient methods, SAC and D4PG .] [Updated on 2018-09-30: ad

We’re launching a transfer learning contest that measures a reinforcement learning algorithm’s ability to generalize from previous experience.

On March 3rd, we hosted our first hackathon with 100 members of the artificial intelligence community.

We’ve developed a simple meta-learning algorithm called Reptile which works by repeatedly sampling a task, performing stochastic gradient descent on it, and upd

We’re providing 6–10 stipends and mentorship to individuals from underrepresented groups to study deep learning full-time for 3 months and open-source a project

Come to OpenAI’s office in San Francisco’s Mission District for talks and a hackathon on Saturday, March 3rd.

We’ve co-authored a paper that forecasts how malicious actors could misuse AI technology, and potential ways we can prevent and mitigate these threats. This pap

[Updated on 2020-09-03: Updated the algorithm of SARSA and Q-learning so that the diffe

We’ve designed a method that encourages AIs to teach each other with examples that also make sense to humans. Our approach automatically selects the most inform

We’re releasing highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights. Depending on the ch

By creating user interfaces which let us work with the representations inside machine learning models, we can give people new tools for reasoning.

We’ve developed a hierarchical reinforcement learning algorithm that learns high-level actions useful for solving a range of tasks, allowing fast solving of tas

Our latest robotics techniques allow robot controllers, trained entirely in simulation and deployed on physical robots, to react to unplanned changes in the env

We’ve found that self-play allows simulated AIs to discover physical skills like tackling, ducking, faking, kicking, catching, and diving for the ball, without

We show that for the task of simulated robot wrestling, a meta-learning agent can learn to quickly defeat a stronger non-meta-learning agent, and also show that

Professor Naftali Tishby passed away in 2021. Hope the post can introduce his cool idea of information bottleneck to more people. Recently I watched the talk &l

We’re releasing an algorithm which accounts for the fact that other agents are learning too, and discovers self-interested yet collaborative strategies like tit

[Updated on 2018-09-30: thanks to Yoonju, we have this post translated in Korean !] [

We’re releasing two new OpenAI Baselines implementations: ACKTR and A2C. A2C is a synchronous, deterministic variant of Asynchronous Advantage Actor Critic (A3C

Our Dota 2 result shows that self-play can catapult the performance of machine learning systems from far below human level to superhuman, given sufficient compu

We’ve created a bot which beats the world’s top professionals at 1v1 matches of Dota 2 under standard tournament rules. The bot learned the game from scratch by

RL-Teacher is an open-source implementation of our interface to train AIs via occasional human feedback rather than hand-crafted reward functions. The underlyin

We’ve found that adding adaptive noise to the parameters of reinforcement learning algorithms frequently boosts performance. This exploration method is simple t

We’re releasing a new class of reinforcement learning algorithms, Proximal Policy Optimization (PPO), which perform comparably or better than state-of-the-art a

One step towards building safe AI systems is to remove the need for humans to write goal functions, since using a simple proxy for a complex goal, or getting th

Multiagent environments where agents compete for resources are stepping stones on the path to AGI. Multiagent environments have two useful properties: first, th

We’re open-sourcing OpenAI Baselines, our internal effort to reproduce reinforcement learning algorithms with performance on par with published results. We’ll r

We’ve created a robotics system, trained entirely in simulation and deployed on a physical robot, which can learn a new task after seeing it done once.

We are releasing Roboschool: open-source software for robot simulation, integrated with OpenAI Gym.

We’ve created the world’s first Spam-detecting AI trained entirely in simulation and deployed on a physical robot.

We’ve discovered that evolution strategies (ES), an optimization technique that’s been known for decades, rivals the performance of standard reinforcement learn

In this post we’ll outline new OpenAI research in which agents develop their own language.

Adversarial examples are inputs to machine learning models that an attacker has intentionally designed to cause the model to make a mistake; they’re like optica

The OpenAI team is now 45 people. Together, we’re pushing the frontier of AI capabilities—whether by validating novel ideas, creating new software systems, or d

Reinforcement learning algorithms can break in surprising, counterintuitive ways. In this post we’ll explore one failure mode, which is where you misspecify you