Concept
OpenAI News
Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by pre-training on a large corpus of text followed by fine-tuning on a specific task. While typically task-agnostic in architecture, this method still requires task-specific...
Topics: ModelsInfrastructure
Entities: ModelsInfrastructure
OpenAI News
We study empirical scaling laws for language model performance on the cross-entropy loss. The loss scales as a power-law with model size, dataset size, and the amount of compute used for training, with some trends spanning more than seven orders of...
Topics: ModelsInfrastructure
Entities: ModelsInfrastructure
OpenAI News
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...
Topics: ModelsInfrastructure
Entities: OpenAIModelsInfrastructure
OpenAI News
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...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
We’re releasing Safety Gym, a suite of environments and tools for measuring progress towards reinforcement learning agents that respect safety constraints while training.
Topics: AgentsInfrastructure
Entities: AgentsInfrastructure
OpenAI News
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...
Topics: AgentsInfrastructure
Entities: AgentsInfrastructure
OpenAI News
We’ve trained a pair of neural networks to solve the Rubik’s Cube with a human-like robot hand. The neural networks are trained entirely in simulation, using the same reinforcement learning code as OpenAI Five paired with a new technique called Automatic...
Topics: ModelsInfrastructure
Entities: OpenAIModelsInfrastructure
OpenAI News
We’ve observed agents discovering progressively more complex tool use while playing a simple game of hide-and-seek. Through training in our new simulated hide-and-seek environment, agents build a series of six distinct strategies and counterstrategies, some...
Topics: AgentsInfrastructure
Entities: AgentsInfrastructure
OpenAI News
We’ve developed a method to assess whether a neural network classifier can reliably defend against adversarial attacks not seen during training. Our method yields a new metric, UAR (Unforeseen Attack Robustness), which evaluates the robustness of a single...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
We’ve made progress towards stable and scalable training of energy-based models (EBMs) resulting in better sample quality and generalization ability than existing models. Generation in EBMs spends more compute to continually refine its answers and doing so...
Topics: ModelsInfrastructure
Entities: ModelsInfrastructure
OpenAI News
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...
Topics: ModelsInfrastructure
Entities: ModelsInfrastructure
OpenAI News
We’ve discovered that the gradient noise scale, a simple statistical metric, predicts the parallelizability of neural network training on a wide range of tasks. Since complex tasks tend to have noisier gradients, increasingly large batch sizes are likely to...
Topics: Infrastructure
Entities: Infrastructure