Backwards State-space Reduction for Planning in Dynamic Knowledge Bases

2014-07-30 · Valerio Senni, Michele Stawowy

In this paper we address the problem of planning in rich domains, where knowledge representation is a key aspect for managing the complexity and size of the planning domain. We follow the approach of Description Logic (DL) based Dynamic Knowledge Bases, where a state of the world is represented concisely by a (possibly changing) ABox and a (fixed) TBox containing the axioms, and actions that allow to change the content of the ABox. The plan goal is given in terms of satisfaction of a DL query. In this paper we start from a traditional forward planning algorithm and we propose a much more efficient variant by combining backward and forward search. In particular, we propose a Backward State-space Reduction technique that consists in two phases: first, an Abstract Planning Graph P is created by using the Abstract Backward Planning Algorithm (ABP), then the abstract planning graph P is instantiated into a corresponding planning graph P by using the Forward Plan Instantiation Algorithm (FPI). The advantage is that in the preliminary ABP phase we produce a symbolic plan that is a pattern to direct the search of the concrete plan. This can be seen as a kind of informed search where the preliminary backward phase is useful to discover properties of the state-space that can be used to direct the subsequent forward phase. We evaluate the effectiveness of our ABP+FPI algorithm in the reduction of the explored planning domain by comparing it to a standard forward planning algorithm and applying both of them to a concrete business case study.

📄 PDF Abstract BibTeX arXiv:1407.7934

Code (0)

등록된 구현이 없습니다.

Task Scoping: Generating Task-Specific Abstractions for Planning in Open-Scope Models

2020-10-17 · Michael Fishman, Nishanth Kumar, Cameron Allen, Natasha Danas 외

A general-purpose planning agent requires an open-scope world model: one rich enough to tackle any of the wide range of tasks it may be asked to solve over its operational lifetime. This stands in contrast with typical p…

Minecraft

Learning retrosynthetic planning through self-play

2019-01-19 · John S. Schreck, Connor W. Coley, Kyle J. M. Bishop

The problem of retrosynthetic planning can be framed as one player game, in which the chemist (or a computer program) works backwards from a molecular target to simpler starting materials though a series of choices regar…

Deep Reinforcement Learning Multi-step retrosynthesis Reinforcement Learning

Differentiable Generalised Predictive Coding

2021-12-02 · André Ofner, Sebastian Stober

This paper deals with differentiable dynamical models congruent with neural process theories that cast brain function as the hierarchical refinement of an internal generative model explaining observations. Our work exten…

Risk Sensitive Model-Based Reinforcement Learning using Uncertainty Guided Planning

2021-11-09 · Stefan Radic Webster, Peter Flach

Identifying uncertainty and taking mitigating actions is crucial for safe and trustworthy reinforcement learning agents, especially when deployed in high-risk environments. In this paper, risk sensitivity is promoted in …

Model-based Reinforcement Learning reinforcement-learning Reinforcement Learning Reinforcement Learning (RL)

Discovery of Dynamics Using Linear Multistep Methods

2019-12-29 · Rachael Keller, Qiang Du

Linear multistep methods (LMMs) are popular time discretization techniques for the numerical solution of differential equations. Traditionally they are applied to solve for the state given the dynamics (the forward probl…