OpenAIAI safety via debate
Exploring AI safety strategies using debate techniques
Jane StreetRepeatable exploratory programming
A blogpost about using Expect tests for repeatable exploratory programming
OpenAIEvolved Policy Gradients
Exploring advanced techniques for policy gradients in machine learning models
OpenAIGotta Learn Fast: A new benchmark for generalization in RL
Exploring advancements in generalization for Reinforcement Learning through a new benchmark approach.
OpenAIRetro Contest
Explore the resurgence of vintage development challenges in the modern era.
Jane StreetOCaml all the way down
Exploring the evolution of the software stack at Jane Street from statistical research to strategy execution.
Jane StreetPutting the I back in IDE: Towards a Github Explorer
A system for editing and reviewing code that puts the I back in IDE.
SoundCloudManaging Unplanned and Support Tasks
A blogpost about a process for managing unplanned and support tasks to minimize interruptions and allocate more time to planned features.
OpenAIVariance reduction for policy gradient with action-dependent factorized baselines
Optimizing policy gradient efficiency with action-dependent factorized baselines
OpenAIImproving GANs using optimal transport
Enhancing GANs performance through optimal transport techniques
OpenAIReport from the OpenAI hackathon
Insights and Highlights from the OpenAI hackathon
OpenAIOn first-order meta-learning algorithms
Exploring the advancements in first-order meta-learning algorithms