Learning to summarize with human feedback

We’ve applied reinforcement learning from human feedback to train language models that are better at summarization.

OpenAI Scholars 2020: Final projects

Our third class of OpenAI Scholars presented their final projects at virtual Demo Day, showcasing their research results from over the past five months.

The Reformer - Pushing the limits of language modeling

Procgen and MineRL Competitions

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.

Image GPT

We find that, just as a large transformer model trained on language can generate coherent text, the same exact model trained on pixel sequences can generate coherent image completions and samples. By establishing a correlation between sample quality and image classification accuracy, we show that o…

OpenAI API

We’re releasing an API for accessing new AI models developed by OpenAI.

Language models are few-shot learners

AI and efficiency

We’re releasing an analysis showing that since 2012 the amount of compute needed to train a neural net to the same performance on ImageNet classification has been decreasing by a factor of 2 every 16 months. Compared to 2012, it now takes 44 times less compute to train a neural network to the level…

Jukebox

We’re introducing Jukebox, a neural net that generates music, including rudimentary singing, as raw audio in a variety of genres and artist styles. We’re releasing the model weights and code, along with a tool to explore the generated samples.

Improving verifiability in AI development

We’ve contributed to a multi-stakeholder report by 58 co-authors at 30 organizations, including the Centre for the Future of Intelligence, Mila, Schwartz Reisman Institute for Technology and Society, Center for Advanced Study in the Behavioral Sciences, and Center for Security and Emerging Technolo…

OpenAI Microscope

We’re introducing OpenAI Microscope, a collection of visualizations of every significant layer and neuron of eight vision “model organisms” which are often studied in interpretability. Microscope makes it easier to analyze the features that form inside these neural networks, and we hope it will hel…

How to generate text: using different decoding methods for language generation with Transformers

How to train a new language model from scratch using Transformers and Tokenizers

OpenAI standardizes on PyTorch

We are standardizing OpenAI’s deep learning framework on PyTorch.

Scaling laws for neural language models

Dota 2 with large scale deep reinforcement learning

Deep double descent

We show that the double descent phenomenon occurs in CNNs, ResNets, and transformers: performance first improves, then gets worse, and then improves again with increasing model size, data size, or training time. This effect is often avoided through careful regularization. While this behavior appear…

Procgen Benchmark

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.

Benchmarking safe exploration in deep reinforcement learning

Safety Gym

We’re releasing Safety Gym, a suite of environments and tools for measuring progress towards reinforcement learning agents that respect safety constraints while training.