Sciweavers

250 search results - page 6 / 50
» Learning action effects in partially observable domains
Sort
View
PAMI
2007
186views more  PAMI 2007»
14 years 9 months ago
Value-Directed Human Behavior Analysis from Video Using Partially Observable Markov Decision Processes
—This paper presents a method for learning decision theoretic models of human behaviors from video data. Our system learns relationships between the movements of a person, the co...
Jesse Hoey, James J. Little
AIIDE
2007
14 years 12 months ago
Learning a Table Soccer Robot a New Action Sequence by Observing and Imitating
Star-Kick is a commercially available and fully automatic table soccer (foosball) robot, which plays table soccer games against human players on a competitive level. One of our re...
Dapeng Zhang 0002, Bernhard Nebel
ECML
2007
Springer
15 years 3 months ago
Policy Gradient Critics
We present Policy Gradient Actor-Critic (PGAC), a new model-free Reinforcement Learning (RL) method for creating limited-memory stochastic policies for Partially Observable Markov ...
Daan Wierstra, Jürgen Schmidhuber
ICML
1995
IEEE
15 years 10 months ago
Learning Policies for Partially Observable Environments: Scaling Up
Partially observable Markov decision processes (pomdp's) model decision problems in which an agent tries to maximize its reward in the face of limited and/or noisy sensor fee...
Michael L. Littman, Anthony R. Cassandra, Leslie P...
ICRA
2007
IEEE
126views Robotics» more  ICRA 2007»
15 years 3 months ago
A formal framework for robot learning and control under model uncertainty
— While the Partially Observable Markov Decision Process (POMDP) provides a formal framework for the problem of robot control under uncertainty, it typically assumes a known and ...
Robin Jaulmes, Joelle Pineau, Doina Precup