AIROMar 12, 2023

The Planner Optimization Problem: Formulations and Frameworks

arXiv:2303.06768v21 citationsh-index: 74
Originality Synthesis-oriented
AI Analysis

This work addresses the lack of consistent problem definitions and software frameworks for learning planning parameter generators, which is an incremental improvement for researchers and practitioners in automated planning.

The paper tackles the challenge of automatically tuning internal parameters of planners conditioned on problem instances by proposing a unified formulation called the planner optimization problem (POP) and an extensible software framework named Open Planner Optimization Framework (OPOF) to specify and solve these problems consistently and reusably.

Identifying internal parameters for planning is crucial to maximizing the performance of a planner. However, automatically tuning internal parameters which are conditioned on the problem instance is especially challenging. A recent line of work focuses on learning planning parameter generators, but lack a consistent problem definition and software framework. This work proposes the unified planner optimization problem (POP) formulation, along with the Open Planner Optimization Framework (OPOF), a highly extensible software framework to specify and to solve these problems in a reusable manner.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes