3 papers
cs.LG2026
Muon: Boosting Muon via Adaptive Second-Moment Preconditioning
Ziyue Liu, Ruijie Zhang, Zhengyang Wang +4
Muon has emerged as a promising optimizer for large-scale foundation model pre-training by exploiting the matrix structure of neural network updates through iterative orthogonaliza…
cs.LG2025
A Unified Contrastive-Generative Framework for Time Series Classification
Ziyu Liu, Azadeh Alavi, Minyi Li +1
Self-supervised learning (SSL) for multivariate time series mainly includes two paradigms: contrastive methods that excel at instance discrimination and generative approaches that…
cs.LG2024
Guidelines for Augmentation Selection in Contrastive Learning for Time Series Classification
Ziyu Liu, Azadeh Alavi, Minyi Li +1
Self-supervised contrastive learning has become a key technique in deep learning, particularly in time series analysis, due to its ability to learn meaningful representations witho…