Learning to Rank for Synthesizing Planning Heuristics

08/03/2016
by   Caelan Reed Garrett, et al.
0

We investigate learning heuristics for domain-specific planning. Prior work framed learning a heuristic as an ordinary regression problem. However, in a greedy best-first search, the ordering of states induced by a heuristic is more indicative of the resulting planner's performance than mean squared error. Thus, we instead frame learning a heuristic as a learning to rank problem which we solve using a RankSVM formulation. Additionally, we introduce new methods for computing features that capture temporal interactions in an approximate plan. Our experiments on recent International Planning Competition problems show that the RankSVM learned heuristics outperform both the original heuristics and heuristics learned through ordinary regression.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/27/2016

Learning and Tuning Meta-heuristics in Plan Space Planning

In recent years, the planning community has observed that techniques for...
research
08/23/2023

Utilizing Admissible Bounds for Heuristic Learning

While learning a heuristic function for forward search algorithms with m...
research
09/27/2011

Improving Heuristics Through Relaxed Search - An Analysis of TP4 and HSP*a in the 2004 Planning Competition

The hm admissible heuristics for (sequential and temporal) regression pl...
research
07/07/2022

Sampling from Pre-Images to Learn Heuristic Functions for Classical Planning

We introduce a new algorithm, Regression based Supervised Learning (RSL)...
research
11/02/2014

A Multi-Heuristic Approach for Solving the Pre-Marshalling Problem

Minimizing the number of reshuffling operations at maritime container te...
research
07/30/2014

Graph Transformation Planning via Abstraction

Modern software systems increasingly incorporate self-* behavior to adap...
research
06/26/2011

SAPA: A Multi-objective Metric Temporal Planner

SAPA is a domain-independent heuristic forward chaining planner that can...

Please sign up or login with your details

Forgot password? Click here to reset