Failure driven dynamic search control for partial order planners: an explanation based approach

Abstract

Given the intractability of domain-independent planning, the ability to control the search of a planner is vitally important. One way of doing this involves learning from search failures. This paper describes SNLP+EBL, the first implementation of explanation based search control rule learning framework for a partial order (plan-space) planner. We will start by describing the basic learning framework of SNLP+EBL. We will then concentrate on SNLP+EBL's ability to learn from failures, and describe the results of empirical studies which demonstrate the effectiveness of the search-control rules SNLP+EBL learns using our method.We then demonstrate the generality of our learning methodology by extending it to UCPOP [39], a descendant of SNLP that allows for more expressive domain theories. The resulting system, UCPOP+EBL, is used to analyze and understand the factors influencing the effectiveness of EBL. Specifically, we analyze the effect of (i) expressive action representations (ii) domain specific failure theories and (iii) sophisticated backtracking strategies on the utility of EBL. Through empirical studies, we demonstrate that expressive action representations allow for more explicit domain representations which in turn increase the ability of EBL to learn from analytical failures, and obviate the need for domain specific failure theories. We also explore the strong affinity between dependency directed backtracking and EBL in planning.

Automatic Induction of Bellman-Error Features for Probabilistic Planning

“…These techniques are related to ours in that both acquire domain specific knowledge via planning experience in the domain. Much of this literature targets control knowledge for particular search-based planners (Estlin & Mooney, 1997;Kambhampati et al, 1996;Veloso et al, 1995), and is distant from our approach in its focus on the particular planning technology used and on the limitation to deterministic domains. It is unclear how to generalize this work to value-function construction or probabilistic domains.…”

Learning Rewrite Rules versus Search Control Rules to Improve Plan Quality

“…Minton's 10] PRODIGY/EBL learned control rules by explaining why a search node leads to success or failure. Kambhampati et al 6] propose a technique based on EBL to learn control rules for partial-order planners and apply it to SNLP and UCPOP to learn rejectionrules. Ihrig et al 4] extended SNLP+EBL to learn from planning successes as well as failures.…”

Storing and Indexing Plan Derivations through Explanation-based Analysis of Retrieval Failures

“…A leaf node plan represents an analytical failure when it contains a set of inconsistent constraints which prevent the plan from being further re ned into a solution. An analytical failure is explained in terms of these constraints (Kambhampati, Katukam, & Qu, 1996b). Leaf node failure explanations identify a minimal set of constraints in the plan which are together inconsistent.…”