Topic feed
AI chips, data centers, training systems, and inference infrastructure.
OpenAI News
We explore methods for option discovery based on variational inference and make two algorithmic contributions. First: we highlight a tight connection between variational option discovery methods and variational autoencoders, and introduce Variational...
Topics: PolicyInfrastructure
Entities: PolicyInfrastructure
OpenAI News
We’ve obtained state-of-the-art results on a suite of diverse language tasks with a scalable, task-agnostic system, which we’re also releasing. Our approach is a combination of two existing ideas: transformers and unsupervised pre-training. These results...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
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-month doubling time (by comparison, Moore’s Law had a 2-year doubling period)[^footnote-correction]....
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
We’re releasing an experimental metalearning approach called Evolved Policy Gradients, a method that evolves the loss function of learning agents, which can enable fast training on novel tasks. Agents trained with EPG can succeed at basic tasks at test time...
Topics: AgentsPolicyInfrastructure
Entities: AgentsPolicyInfrastructure
OpenAI News
This paper considers meta-learning problems, where there is a distribution of tasks, and we would like to obtain an agent that performs well (i.e., learns quickly) when presented with a previously unseen task sampled from this distribution. We analyze a...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
We’re releasing highly-optimized GPU kernels for an underexplored class of neural network architectures: networks with block-sparse weights. Depending on the chosen sparsity, these kernels can run orders of magnitude faster than cuBLAS or cuSPARSE. We’ve...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
We propose a practical method for L₀ norm regularization for neural networks: pruning the network during training by encouraging weights to become exactly zero. Such regularization is interesting since (1) it can greatly speed up training and inference, and...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
Teachers intentionally pick the most informative examples to show their students. However, if the teacher and student are neural networks, the examples that the teacher network learns to give, although effective at teaching the student, are typically...
Topics: Infrastructure
Entities: Infrastructure
OpenAI News
Simulations are attractive environments for training agents as they provide an abundant source of data and alleviate certain safety concerns during the training process. But the behaviours developed by agents in simulation are often specific to the...
Topics: AgentsInfrastructure
Entities: AgentsInfrastructure
OpenAI News
Deep reinforcement learning (RL) has proven a powerful technique in many sequential decision making domains. However, Robotics poses many challenges for RL, most notably training on a physical system can be expensive and dangerous, which has sparked...
Topics: PolicyInfrastructure
Entities: PolicyInfrastructure
OpenAI News
Multi-agent settings are quickly gathering importance in machine learning. This includes a plethora of recent work on deep multi-agent reinforcement learning, but also can be extended to hierarchical RL, generative adversarial networks and decentralised...
Topics: AgentsPolicyInfrastructure
Entities: AgentsPolicyInfrastructure
OpenAI News
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 compute. In the span of a month, our system went from barely matching a high-ranked player to beating...
Topics: Infrastructure
Entities: Infrastructure