Open Peer Review. Open Publishing. Open Access. Open Discussion. Open Directory. Open Recommendations. Open API. Open Source.
An Analysis of Feature Regularization for Low-shot Learning
Zhuoyuan Chen, Han Zhao, Xiao Liu, Wei Xu
Nov 04, 2016 (modified: Jan 11, 2017)ICLR 2017 conference submissionreaders: everyone
Abstract:Low-shot visual learning, the ability to recognize novel object categories from very few, or even one example, is a hallmark of human visual intelligence. Though successful on many tasks, deep learning approaches tends to be notoriously data-hungry. Recently, feature penalty regularization has been proved effective on capturing new concepts. In this work, we provide both empirical evidence and theoretical analysis on how and why these methods work. We also propose a better design of cost function with improved performance. Close scrutiny reveals the centering effect of feature representation, as well as the intrinsic connection with batch normalization. Extensive experiments on synthetic datasets, the one-shot learning benchmark “Omniglot”, and large-scale ImageNet validate our analysis.
TL;DR:An analysis of adding regularization for low-shot learning
Keywords:Deep learning, Computer vision
Enter your feedback below and we'll get back to you as soon as possible.