2 papers
cs.LG2025
Some theoretical improvements on the tightness of PAC-Bayes risk certificates for neural networks
Diego García-Pérez, Emilio Parrado-Hernández, John Shawe-Taylor
This paper presents four theoretical contributions that improve the usability of risk certificates for neural networks based on PAC-Bayes bounds. First, two bounds on the KL diverg…
cs.LG2025
CRITS: Convolutional Rectifier for Interpretable Time Series Classification
Alejandro Kuratomi, Zed Lee, Guilherme Dinis Chaliane Junior +2
Several interpretability methods for convolutional network-based classifiers exist. Most of these methods focus on extracting saliency maps for a given sample, providing a local ex…