Badly trained policy after 40000 steps

Multi-agent DQN training step 90000 trajectory video

Multi-agent DQN training step 0 trajectory video

Learning to grab with bell as reward

Intel Deep Learning Community of Practice talk

Natasha Jaques PhD Thesis Defense

Personalized Multi-task Learning for Predicting Tomorrow's Mood, Stress, and Health

VHRED Cornell baseline

Influence agent in Harvest game

A3C baseline in Harvest

Agent trained with intrinsic social influence reward - Tragedy of the Commons

Agent trained with intrinsic social influence reward

A3C will not free other agent trapped in a box

Influence agent frees compatriot trapped in a box

Note RNN

Q

G

Basic LSTM

Psi

RL Tuner

EDAExplorer PeakTutorial

EDAExplorer ArtifactTutorial

The Challenge

Affective Computing - Spring 2015 Virtual Visit

Eye gaze data

5 Lego Robots Dancing to Gangnam Style