OpenAIOn the quantitative analysis of decoder-based generative models
Exploring the quantitative analysis of decoder-based generative models
OpenAIA connection between generative adversarial networks, inverse reinforcement learning, and energy-based models
Exploring the synergy of generative adversarial networks, inverse reinforcement learning, and energy-based models
OpenAIRL²: Fast reinforcement learning via slow reinforcement learning
Optimizing reinforcement learning with RL² approach
OpenAIVariational lossy autoencoder
Exploring the variational lossy autoencoder for enhanced compression and representation learning.
Jane StreetA solution to the ppx versioning problem
A solution to versioning problem in ppx, a preprocessing system for OCaml that maps over the AST to generate code from special syntax fragments.
OpenAIExtensions and limitations of the neural GPU
Exploring the implications of extending and constraining the neural GPU model
Jane StreetObservations of a functional programmer
A blogpost about observations made by a functional programmer during their keynote at a functional programming workshop.
OpenAISemi-supervised knowledge transfer for deep learning from private training data
Exploring semi-supervised knowledge transfer in deep learning using private training data
OpenAIReport from the self-organizing conference
Exploring the dynamics of a self-organizing conference
OpenAITransfer from simulation to real world through learning deep inverse dynamics model
Optimizing real-world implementation with deep inverse dynamics model
SoundCloudAPI Sign-up Changes
Launching a new API application process that requires developers to fill out an application form to request access.
Jane StreetWhat the interns have wrought, 2016
Reflecting on the impact of the interns on various aspects of the code-base across multiple office locations.