Andi Peng—A Human-in-the-Loop Framework for Test-Time Policy Adaptation
Andi is a ML PhD at MIT, interested in building agents that learn continuously from and with humans. To that end, she spends a lot of time thinking about how to align algorithmic representations with humans’, whether that be through designing novel learning algorithms or questioning the theoretical foundations of agency, goals, and planning. Andy gets most excited by work that unifies RL, robotics, and cognitive science.
Paper: https://arxiv.org/pdf/2307.06333.pdf
Andi: / theandipenguin
Host: / michaeltrazzi
Patreon: / theinsideview
Patreon supporters:
Tassilo Neubauer
MonikerEpsilon
Alexey Malafeev
Jack Seroy
JJ Hepburn
Max Chiswick
William Freire
Edward Huff
Gunnar Höglund
Ryan Coppolo
Cameron Holmes
Emil Wallner
Jesse Hoogland
Jacques Thibodeau
Vincent Weisser