Generalized Value Iteration Networks: Life Beyond Lattices

TL;DR

This paper presents GVIN, a neural network planning module that generalizes value iteration to irregular graphs using novel graph convolution kernels, improving planning on diverse and unseen graph structures.

Contribution

Introduction of GVIN with three novel graph convolution kernels and episodic Q-learning for stable training on irregular and large-scale graphs.

Findings

GVIN outperforms naive VIN on irregular and real-world graphs.

Embedding-based kernel achieves best performance among proposed kernels.

GVIN generalizes well to unseen and larger graphs.

Abstract

In this paper, we introduce a generalized value iteration network (GVIN), which is an end-to-end neural network planning module. GVIN emulates the value iteration algorithm by using a novel graph convolution operator, which enables GVIN to learn and plan on irregular spatial graphs. We propose three novel differentiable kernels as graph convolution operators and show that the embedding based kernel achieves the best performance. We further propose episodic Q-learning, an improvement upon traditional n-step Q-learning that stabilizes training for networks that contain a planning module. Lastly, we evaluate GVIN on planning problems in 2D mazes, irregular graphs, and real-world street networks, showing that GVIN generalizes well for both arbitrary graphs and unseen graphs of larger scale and outperforms a naive generalization of VIN (discretizing a spatial graph into a 2D image).