activity
20172024
most citedLocal Aggregation for Unsupervised Learning of Visual Embeddings

78 citations · 175 across the 5 of their papers we have counts for

collaborators

8 papers

cs.CV2024

The BabyView dataset: High-resolution egocentric videos of infants' and young children's everyday experiences

Bria Long, Robert Z. Sparks, Violet Xiang +9

Human children far exceed modern machine learning algorithms in their sample efficiency, achieving high performance in key domains with much less data than current models. This ''d…

cs.LG202026 cited

Conditional Negative Sampling for Contrastive Learning of Visual Representations

Mike Wu, Milan Mosse, Chengxu Zhuang +2

Recent methods for learning unsupervised visual representations, dubbed contrastive learning, optimize the noise-contrastive estimation (NCE) bound on mutual information between tw…

cs.LG202049 cited

On Mutual Information in Contrastive Learning for Visual Representations

Mike Wu, Chengxu Zhuang, Milan Mosse +2

In recent years, several unsupervised, "contrastive" learning algorithms in vision have been shown to learn representations that perform remarkably well on transfer tasks. We show…

cs.CV20197 cited

Local Label Propagation for Large-Scale Semi-Supervised Learning

Chengxu Zhuang, Xuehao Ding, Divyanshu Murli +1

A significant issue in training deep neural networks to solve supervised learning tasks is the need for large numbers of labelled datapoints. The goal of semi-supervised learning i…

cs.CV2019

Unsupervised Learning from Video with Deep Neural Embeddings

Chengxu Zhuang, Tianwei She, Alex Andonian +2

Because of the rich dynamical structure of videos and their ubiquity in everyday life, it is a natural idea that video data could serve as a powerful unsupervised learning signal f…

cs.CV201978 cited

Local Aggregation for Unsupervised Learning of Visual Embeddings

Chengxu Zhuang, Alex Lin Zhai, Daniel Yamins

Unsupervised approaches to learning in neural networks are of substantial interest for furthering artificial intelligence, both because they would enable the training of networks w…