AI · Education
The power of continuous learning
Lilian Weng works on Applied AI Research at OpenAI.
Dec 23 · 08:00OpenAI
1,690 press releases in English indexed under education.
Lilian Weng works on Applied AI Research at OpenAI.
Abstract In reinforcement learning from human feedback, it is common to optimize against a reward model trained to predict human preferences. Because the reward model is an imperfect proxy, optimizing its value too much can hinder ground truth performance, in accordance with Goodhart's law. This effect has been frequently observed, but not carefully measured due to the expense of collecting human preference data. In
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 computer-using agents.
We built a neural theorem prover for Lean that learned to solve a variety of challenging high-school olympiad problems, including problems from the AMC12 and AIME competitions, as well as two problems adapted from the IMO.
We’ve trained a system that solves grade school math problems with nearly twice the accuracy of a fine-tuned GPT-3 model. It solves about 90% as many problems as real kids: a small sample of 9-12 year olds scored 60% on a test from our dataset, while our system scored 55% on those same problems.
We’ve applied reinforcement learning from human feedback to train language models that are better at summarization.
We’re excited to announce that OpenAI is co-organizing two NeurIPS 2020 competitions with AIcrowd, Carnegie Mellon University, and DeepMind, using Procgen Benchmark and MineRL.
We are standardizing OpenAI’s deep learning framework on PyTorch.
Abstract On April 13th, 2019, OpenAI Five became the first AI system to defeat the world champions at an esports game. The game of Dota 2 presents novel challenges for AI systems such as long time horizons, imperfect information, and complex, continuous state-action spaces, all challenges which will become increasingly central to more capable AI systems. OpenAI Five leveraged existing reinforcement learning technique
We’re releasing Procgen Benchmark, 16 simple-to-use procedurally-generated environments which provide a direct measure of how quickly a reinforcement learning agent learns generalizable skills.
We’re releasing Safety Gym, a suite of environments and tools for measuring progress towards reinforcement learning agents that respect safety constraints while training.
Abstract Reinforcement learning (RL) agents need to explore their environments in order to learn optimal policies by trial and error. In many environments, safety is a critical concern and certain errors are unacceptable: for example, robotics systems that interact with humans should never cause injury to the humans while exploring. While it is currently typical to train RL agents mostly or entirely in simulation, wh