PixelCNN++: Improving the PixelCNN with discretized logistic mixture likelihood and other modifications
AI Labs
AI Labs News
Announcements from OpenAI, Google, Anthropic, Hugging Face and other labs
Faulty reward functions in the wild
AI Labs
Universe
AI Labs
#Exploration: A study of count-based exploration for deep reinforcement learning
AI Labs
OpenAI and Microsoft
AI Labs
On the quantitative analysis of decoder-based generative models
AI Labs
A connection between generative adversarial networks, inverse reinforcement learning, and energy-based models
AI Labs
RL²: Fast reinforcement learning via slow reinforcement learning
AI Labs
Variational lossy autoencoder
AI Labs
Extensions and limitations of the neural GPU
AI Labs
Semi-supervised knowledge transfer for deep learning from private training data
AI Labs
Report from the self-organizing conference
AI Labs