Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Fast Tree-Field Integrators: From Low Displacement Rank to Topological Transformers

About

We present a new class of fast polylog-linear algorithms based on the theory of structured matrices (in particular low displacement rank) for integrating tensor fields defined on weighted trees. Several applications of the resulting fast tree-field integrators (FTFIs) are presented, including (a) approximation of graph metrics with tree metrics, (b) graph classification, (c) modeling on meshes, and finally (d) Topological Transformers (TTs) (Choromanski et al., 2022) for images. For Topological Transformers, we propose new relative position encoding (RPE) masking mechanisms with as few as three extra learnable parameters per Transformer layer, leading to 1.0-1.5%+ accuracy gains. Importantly, most of FTFIs are exact methods, thus numerically equivalent to their brute-force counterparts. When applied to graphs with thousands of nodes, those exact algorithms provide 5.7-13x speedups. We also provide an extensive theoretical analysis of our methods.

Krzysztof Choromanski, Arijit Sehanobish, Somnath Basu Roy Chowdhury, Han Lin, Avinava Dubey, Tamas Sarlos, Snigdha Chaturvedi• 2024

Related benchmarks

TaskDatasetResultRank
Graph ClassificationPROTEINS
Accuracy72.5
1383
Graph ClassificationMUTAG
Accuracy82.2
1229
Graph ClassificationNCI1
Accuracy73.7
707
Graph ClassificationCOLLAB
Accuracy75.5
532
Graph ClassificationENZYMES
Accuracy42.5
419
Graph ClassificationPTC-MR
Accuracy60.6
271
Graph ClassificationD&D
Accuracy74.8
179
Graph ClassificationIMDB MULTI
Accuracy47.6
168
Graph Classificationimdb-binary
Accuracy65.1
156
Graph ClassificationREDDIT BINARY
Accuracy84.3
144
Showing 10 of 14 rows

Other info

Code

Follow for update