Journal Title
Title of Journal: Appl Intell
|
Abbravation: Applied Intelligence
|
|
|
|
|
|
|
|
Authors: Ngo Anh Vien SeungGwan Lee TaeChoong Chung
Publish Date: 2016/01/29
Volume: 45, Issue: 1, Pages: 112-126
Abstract
Reinforcement learning RL is an area of machine learning that is concerned with how an agent learns to make decisions sequentially in order to optimize a particular performance measure For achieving such a goal the agent has to choose either 1 exploiting previously known knowledge that might end up at local optimality or 2 exploring to gather new knowledge that expects to improve the current performance Among other RL algorithms Bayesian modelbased RL BRL is wellknown to be able to tradeoff between exploitation and exploration optimally via belief planning ie partially observable Markov decision process POMDP However solving that POMDP often suffers from curse of dimensionality and curse of history In this paper we make two major contributions which are 1 an integration framework of temporal abstraction into BRL that eventually results in a hierarchical POMDP formulation which can be solved online using a hierarchical samplebased planning solver 2 a subgoal discovery method for hierarchical BRL that automatically discovers useful macro actions to accelerate learning In the experiment section we demonstrate that the proposed approach can scale up to much larger problems On the other hand the agent is able to discover useful subgoals for speeding up Bayesian reinforcement learningThe authors are grateful to the Basic Science Research Program through the National Research Foundation of Korea NRF funded by the Ministry of Education Science and Technology 2014R1A1A2057735 for its tremendous support to this work’s completion
Keywords:
.
|
Other Papers In This Journal:
|