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

3 results for “L2-Normalization”

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 ↗

Virtual Adversarial Ladder Networks For Semi-supervised Learningcs.LG

Semi-supervised learning (SSL) partially circumvents the high cost of labeling data by augmenting a small labeled dataset with a large and relatively cheap unlabeled dataset drawn from the same distribution. This paper offers a novel interpretation of two deep learning-based SSL approaches, ladder networks and virtual …

2017-11-20abs ↗pdf ↗

A Spherical Hidden Markov Model for Semantics-Rich Human Mobility Modelingcs.LG

SHMM models human mobility from GPS and text data, overcoming text sparsity.

problem Modeling human mobility from semantic trace data, especially addressing text sparsity.

method SHMM is a multi-modal spherical hidden Markov model that jointly models location, time, and text embeddings on a unit sphere using vMF distribution.

result SHMM outperforms state-of-the-art models in next location prediction and has lower training cost.

human mobility semantics-rich spherical hidden Markov model text embeddings next location prediction

2020-10-05abs ↗pdf ↗