Bridging Commonsense Reasoning and Probabilistic Planning via a Probabilistic Action Language

Y. I. Wang, Shiqi Zhang, Joohyung Lee

Research output: Contribution to journalArticle

Abstract

To be responsive to dynamically changing real-world environments, an intelligent agent needs to perform complex sequential decision-making tasks that are often guided by commonsense knowledge. The previous work on this line of research led to the framework called interleaved commonsense reasoning and probabilistic planning (icorpp), which used P-log for representing commmonsense knowledge and Markov Decision Processes (MDPs) or Partially Observable MDPs (POMDPs) for planning under uncertainty. A main limitation of icorpp is that its implementation requires non-trivial engineering efforts to bridge the commonsense reasoning and probabilistic planning formalisms. In this paper, we present a unified framework to integrate icorpp's reasoning and planning components. In particular, we extend probabilistic action language pBC+ to express utility, belief states, and observation as in POMDP models. Inheriting the advantages of action languages, the new action language provides an elaboration tolerant representation of POMDP that reflects commonsense knowledge. The idea led to the design of the system pbcplus2pomdp, which compiles a pBC+ action description into a POMDP model that can be directly processed by off-the-shelf POMDP solvers to compute an optimal policy of the pBC+ action description. Our experiments show that it retains the advantages of icorpp while avoiding the manual efforts in bridging the commonsense reasoner and the probabilistic planner.

Original languageEnglish (US)
Pages (from-to)1090-1106
Number of pages17
JournalTheory and Practice of Logic Programming
Volume19
Issue number5-6
DOIs
StatePublished - Sep 1 2019

Fingerprint

Reasoning
Planning
Partially Observable Markov Decision Process
Intelligent agents
Intelligent Agents
Markov Decision Process
Optimal Policy
Language
Express
Decision making
Decision Making
Integrate
Engineering
Uncertainty
Line
Model
Experiment
Knowledge
Experiments
Framework

Keywords

  • Action Language
  • Commonsense Reasoning
  • POMDP
  • Probabilistic Logic Programming
  • Probabilistic Planning

ASJC Scopus subject areas

  • Software
  • Theoretical Computer Science
  • Hardware and Architecture
  • Computational Theory and Mathematics
  • Artificial Intelligence

Cite this

Bridging Commonsense Reasoning and Probabilistic Planning via a Probabilistic Action Language. / Wang, Y. I.; Zhang, Shiqi; Lee, Joohyung.

In: Theory and Practice of Logic Programming, Vol. 19, No. 5-6, 01.09.2019, p. 1090-1106.

Research output: Contribution to journalArticle

@article{bd914e3a5ce949d38d778395c2be4194,
title = "Bridging Commonsense Reasoning and Probabilistic Planning via a Probabilistic Action Language",
abstract = "To be responsive to dynamically changing real-world environments, an intelligent agent needs to perform complex sequential decision-making tasks that are often guided by commonsense knowledge. The previous work on this line of research led to the framework called interleaved commonsense reasoning and probabilistic planning (icorpp), which used P-log for representing commmonsense knowledge and Markov Decision Processes (MDPs) or Partially Observable MDPs (POMDPs) for planning under uncertainty. A main limitation of icorpp is that its implementation requires non-trivial engineering efforts to bridge the commonsense reasoning and probabilistic planning formalisms. In this paper, we present a unified framework to integrate icorpp's reasoning and planning components. In particular, we extend probabilistic action language pBC+ to express utility, belief states, and observation as in POMDP models. Inheriting the advantages of action languages, the new action language provides an elaboration tolerant representation of POMDP that reflects commonsense knowledge. The idea led to the design of the system pbcplus2pomdp, which compiles a pBC+ action description into a POMDP model that can be directly processed by off-the-shelf POMDP solvers to compute an optimal policy of the pBC+ action description. Our experiments show that it retains the advantages of icorpp while avoiding the manual efforts in bridging the commonsense reasoner and the probabilistic planner.",
keywords = "Action Language, Commonsense Reasoning, POMDP, Probabilistic Logic Programming, Probabilistic Planning",
author = "Wang, {Y. I.} and Shiqi Zhang and Joohyung Lee",
year = "2019",
month = "9",
day = "1",
doi = "10.1017/S1471068419000371",
language = "English (US)",
volume = "19",
pages = "1090--1106",
journal = "Theory and Practice of Logic Programming",
issn = "1471-0684",
publisher = "Cambridge University Press",
number = "5-6",

}

TY - JOUR

T1 - Bridging Commonsense Reasoning and Probabilistic Planning via a Probabilistic Action Language

AU - Wang, Y. I.

AU - Zhang, Shiqi

AU - Lee, Joohyung

PY - 2019/9/1

Y1 - 2019/9/1

N2 - To be responsive to dynamically changing real-world environments, an intelligent agent needs to perform complex sequential decision-making tasks that are often guided by commonsense knowledge. The previous work on this line of research led to the framework called interleaved commonsense reasoning and probabilistic planning (icorpp), which used P-log for representing commmonsense knowledge and Markov Decision Processes (MDPs) or Partially Observable MDPs (POMDPs) for planning under uncertainty. A main limitation of icorpp is that its implementation requires non-trivial engineering efforts to bridge the commonsense reasoning and probabilistic planning formalisms. In this paper, we present a unified framework to integrate icorpp's reasoning and planning components. In particular, we extend probabilistic action language pBC+ to express utility, belief states, and observation as in POMDP models. Inheriting the advantages of action languages, the new action language provides an elaboration tolerant representation of POMDP that reflects commonsense knowledge. The idea led to the design of the system pbcplus2pomdp, which compiles a pBC+ action description into a POMDP model that can be directly processed by off-the-shelf POMDP solvers to compute an optimal policy of the pBC+ action description. Our experiments show that it retains the advantages of icorpp while avoiding the manual efforts in bridging the commonsense reasoner and the probabilistic planner.

AB - To be responsive to dynamically changing real-world environments, an intelligent agent needs to perform complex sequential decision-making tasks that are often guided by commonsense knowledge. The previous work on this line of research led to the framework called interleaved commonsense reasoning and probabilistic planning (icorpp), which used P-log for representing commmonsense knowledge and Markov Decision Processes (MDPs) or Partially Observable MDPs (POMDPs) for planning under uncertainty. A main limitation of icorpp is that its implementation requires non-trivial engineering efforts to bridge the commonsense reasoning and probabilistic planning formalisms. In this paper, we present a unified framework to integrate icorpp's reasoning and planning components. In particular, we extend probabilistic action language pBC+ to express utility, belief states, and observation as in POMDP models. Inheriting the advantages of action languages, the new action language provides an elaboration tolerant representation of POMDP that reflects commonsense knowledge. The idea led to the design of the system pbcplus2pomdp, which compiles a pBC+ action description into a POMDP model that can be directly processed by off-the-shelf POMDP solvers to compute an optimal policy of the pBC+ action description. Our experiments show that it retains the advantages of icorpp while avoiding the manual efforts in bridging the commonsense reasoner and the probabilistic planner.

KW - Action Language

KW - Commonsense Reasoning

KW - POMDP

KW - Probabilistic Logic Programming

KW - Probabilistic Planning

UR - http://www.scopus.com/inward/record.url?scp=85072632041&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072632041&partnerID=8YFLogxK

U2 - 10.1017/S1471068419000371

DO - 10.1017/S1471068419000371

M3 - Article

AN - SCOPUS:85072632041

VL - 19

SP - 1090

EP - 1106

JO - Theory and Practice of Logic Programming

JF - Theory and Practice of Logic Programming

SN - 1471-0684

IS - 5-6

ER -