Timezone: »

 
Poster
Approximation Ratios of Graph Neural Networks for Combinatorial Problems
Ryoma Sato · Makoto Yamada · Hisashi Kashima

Tue Dec 10 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #214
in Theory -- Hardness of Learning and Approximations »

In this paper, from a theoretical perspective, we study how powerful graph neural networks (GNNs) can be for learning approximation algorithms for combinatorial problems. To this end, we first establish a new class of GNNs that can solve a strictly wider variety of problems than existing GNNs. Then, we bridge the gap between GNN theory and the theory of distributed local algorithms. We theoretically demonstrate that the most powerful GNN can learn approximation algorithms for the minimum dominating set problem and the minimum vertex cover problem with some approximation ratios with the aid of the theory of distributed local algorithms. We also show that most of the existing GNNs such as GIN, GAT, GCN, and GraphSAGE cannot perform better than with these ratios. This paper is the first to elucidate approximation ratios of GNNs for combinatorial problems. Furthermore, we prove that adding coloring or weak-coloring to each node feature improves these approximation ratios. This indicates that preprocessing and feature engineering theoretically strengthen model capabilities.

Keywords: Deep Learning · Hardness of Learning and Approximations · Theory
[ Paper]  [ Poster] 

Author Information

Ryoma Sato (Kyoto University)

I am a first year student in master's program at Kashima-Yamada Lab, Kyoto University Interests: Machine Learning on Graphs, Discrete Algorithms

Makoto Yamada (Kyoto University/RIKEN AIP)
Hisashi Kashima (Kyoto University/RIKEN Center for AIP)

More from the Same Authors