Neural Weighted A*: Learning Graph Costs and Heuristics with Differentiable Anytime A*

TL;DR

Neural Weighted A* is a differentiable planner that learns graph costs and heuristics from images, allowing end-to-end training and adjustable trade-offs between accuracy and efficiency, outperforming baselines.

Contribution

It introduces Neural Weighted A* that jointly learns costs and heuristics with a differentiable A* solver, enabling end-to-end training and runtime trade-offs.

Findings

Outperforms baselines in planning accuracy and efficiency

Successfully learns graph representations from raw images

Provides controllable suboptimality bounds

Abstract

Recently, the trend of incorporating differentiable algorithms into deep learning architectures arose in machine learning research, as the fusion of neural layers and algorithmic layers has been beneficial for handling combinatorial data, such as shortest paths on graphs. Recent works related to data-driven planning aim at learning either cost functions or heuristic functions, but not both. We propose Neural Weighted A*, a differentiable anytime planner able to produce improved representations of planar maps as graph costs and heuristics. Training occurs end-to-end on raw images with direct supervision on planning examples, thanks to a differentiable A* solver integrated into the architecture. More importantly, the user can trade off planning accuracy for efficiency at run-time, using a single, real-valued parameter. The solution suboptimality is constrained within a linear bound equal to the optimal path cost multiplied by the tradeoff parameter. We experimentally show the validity of our claims by testing Neural Weighted A* against several baselines, introducing a novel, tile-based navigation dataset. We outperform similar architectures in planning accuracy and efficiency.