Learning to Plan Reading Group

Schedule for Fall 2007
I like to cover
Markov Logic Network (will be extended to Stochastic EBL)
Borrajo and Veloso's works
Machine Learning for Oversubscribed Planning and Temporal Planning
Machine Learning for Linear or Integer Programming.

List of Potential Papers for Spring 2007
Concurrent Hierarchical Reinforcement Learning
Huang, Selman and Kautz's work
Learning to take Actions
Max Margin Planning
TLPlan
Learning Evaluation Functions to" Improve Optimization by Local Search
Feature Construction for Reinforcement Learning in Hearts
Improving Heuristic Mini-Max Search by Supervised Learning
Bayesian Inverse Reinforcement Learning
Markov Logic Network
ProbLog: A Probabilistic Prolog and its Application in Link Discovery
Hierarchical reinforcement learning with the MAXQ value function decomposition
Policy Improvement for POMDPs Using Normalized Importance Sampling

Veloso and Borrajo's works
DISTILL: Towards learning domain-specific planners by example,
Multiagent systems: A survey from a machine learning perspective
Lazy incremental learning of control knowledge for efficiently obtaining quality plans
The need for different domain-independent heuristics
　

Fall 2006

I would like to list some reading list.

This list is not meant to be complete.

Any suggestion to the reading list would be appreciated. Please send email to Sungwook.Yoon@asu.edu

Using Temporal Logics to Express Search Control Knowledge for Planning, by Bacchus and Kabanza

SHOP: Simple Hierarchical Ordered Planner by Nau, Cao, Lotem, Muñoz-Avila
- Ultimately we would like to automatically find control knowledge as powerful as these

Learning-assisted automated planning: Looking back, taking stock, going forward by Zimmerman and Kambhampati
- Excellent and Comprehensive List, I suggest read this after understanding basics

FOIL
- Well known ILP system, good candidate for learning control knowledge of relational AI Planning

Exploiting First-Order Regression in Inductive Policy Selection by Gretton and Thiébaux
- To me, there is some EBL flavor in this work and like to develop this work to more concrete EBL

Learning Goal-Decomposition Rules using Exercises by Reddy and Tadepalli
- How to learn gradually harder problems

An Analytic Learning System for Specializing Heuristics by Minton
- EBL

Lazy Incremental Learning of Control Knowledge for Efficiently Obtaining Quality Plans by Borrajo and Veloso

Explanation Based Learning - A Survey of Programs and Perspectives by Ellman
- Another EBL, a nice survey

Learning to Take Actions by Khardon
- Pioneering paper for the modern control knowledge learning problem

Learning Evaluation Functions to" Improve Optimization by Local Search by Boyan and Moore
- Learning to Search paper. Learns additional value function from search experience

Explanation-Based Learning: An Alternative View by Dejong and Mooney
- Just making up explanations is not enough

Relational Reinforcement Learning by Bunch of People
- A workshop summary

Learning Declarative Control Rules for Constraint-Based Planning by Huang, Selman and Kautz
- Finds rules for SAT Planner

Learning generalized policies in planning using concept languages, by Martin and Geffner
- Excellent perspective provided by Hector Geffner

Home