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.

168,657 papers · 148 categories

Trend · papers per month

count % of corpus

2 results for “co-supervision”

SynthTree: Co-supervised Local Model Synthesis for Explainable Predictionstat.ME

Enhances explainability of AI models without sacrificing accuracy.

problem Lack of interpretability in black-box models like Deep Neural Networks and Gradient Boosting.

method Co-supervised Local Model Synthesis (SynthTree) using Mixture of Linear Models (MLM).

result Statistical models significantly enhance explainability of AI models.

explainable AI black-box models Mixture of Linear Models co-supervision agglomerative clustering

2024-06-16abs ↗pdf ↗

Mixture of Linear Models Co-supervised by Deep Neural Networkscs.LG

A new method combines DNN with simpler models for better interpretability and accuracy.

problem Combining interpretability and accuracy in machine learning models.

method A mixture of linear models guided by deep neural networks.

result Improved interpretability without sacrificing accuracy.

deep learning interpretability mixture models neural networks statistical models

2021-08-05abs ↗pdf ↗