On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “Soft-Selector”

Improving Graph Neural Networks with Simple Architecture Designstat.ML

Improved graph neural networks by separating feature aggregation and depth.

problem Understanding feature importance in graph neural networks without prior information.

method Decoupling feature aggregation and depth, using softmax as a regularizer, and introducing 'Soft-Selector' and 'Hop-Normalization'.

result FSGNN model achieves up to 64% accuracy improvements in node classification tasks.

Graph Neural Networks Feature Selection Softmax Regularization Hop-Normalization Soft-Selector

2021-05-17abs ↗pdf ↗

Simplifying approach to Node Classification in Graph Neural Networksstat.ML

Simplifies GNN models by selecting important features for node classification.

problem Challenges in analyzing and selecting important features in GNN models.

method Decoupling feature aggregation and depth, using softmax and L2-normalization.

result FSGNN achieves comparable or higher accuracy than state-of-the-art GNN models.

Graph Neural Networks Node Classification Feature Selection Softmax L2-Normalization

2021-11-12abs ↗pdf ↗