Minimizing Regret


Google Princeton AI and Hazan Lab @ Princeton University


Adaptive Regret for Control of Time-Varying Dynamics

Visualization of the AdaGPC algorithm vs a planning algorithm (iLQR) on the inverted pendulum (Paula Gradu and Edgar Minasyan, paper).

Provably Efficient Maximum Entropy Exploration

Visualization of the MaxEnt exploration algorithm (Karan Singh and Abby Van Soest) from this paper.

YouTube Playlist of talks on optimization/control

Link to the full playlist here.

Share on: