46 citations · 177 across the 41 of their papers we have counts for
4 papers · 1 filter
Decoupled Contrastive Learning
Chun-Hsiao Yeh, Cheng-Yao Hong, Yen-Chi Hsu +3
Contrastive learning (CL) is one of the most successful paradigms for self-supervised learning (SSL). In a principled way, it considers two augmented "views" of the same image as p…
Clipped Hyperbolic Classifiers Are Super-Hyperbolic Classifiers
Yunhui Guo, Xudong Wang, Yubei Chen +1
Hyperbolic space can naturally embed hierarchies, unlike Euclidean space. Hyperbolic Neural Networks (HNNs) exploit such representational power by lifting Euclidean features into h…
Compact and Optimal Deep Learning with Recurrent Parameter Generators
Jiayun Wang, Yubei Chen, Stella X. Yu +2
Deep learning has achieved tremendous success by training increasingly large models, which are then compressed for practical deployment. We propose a drastically different approach…
Transformer visualization via dictionary learning: contextualized embedding as a linear superposition of transformer factors
Zeyu Yun, Yubei Chen, Bruno A Olshausen +1
Transformer networks have revolutionized NLP representation learning since they were introduced. Though a great effort has been made to explain the representation in transformers,…