Maximum Entropy Fine-Grained Classification
Name
7344-maximum-entropy-fine-grained-classification.pdf
Description
Published version
Size
1.24 MB
Format
Adobe PDF
Checksum (MD5)
97e3c69ac6b1fccbbf6f8d0e3b46bc67
Author(s)
Dubey, Abhimanyu
Gupta, Otkrist
Raskar, Ramesh
Naik, Nikhil
Date Issued
2018
Citation
Dubey, Abhimanyu, Gupta, Otkrist, Raskar, Ramesh and Naik, Nikhil. 2018. "Maximum Entropy Fine-Grained Classification."
Version
Final published version
Abstract
© 2018 Curran Associates Inc..All rights reserved. Fine-Grained Visual Classification (FGVC) is an important computer vision problem that involves small diversity within the different classes, and often requires expert annotators to collect data. Utilizing this notion of small visual diversity, we revisit Maximum-Entropy learning in the context of fine-grained classification, and provide a training routine that maximizes the entropy of the output probability distribution for training convolutional neural networks on FGVC tasks. We provide a theoretical as well as empirical justification of our approach, and achieve state-of-the-art performance across a variety of classification tasks in FGVC, that can potentially be extended to any fine-tuning task. Our method is robust to different hyperparameter values, amount of training data and amount of training label noise and can hence be a valuable tool in many similar problems.
MIT Department
Massachusetts Institute of Technology. Media Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
