An Efficient Alternative of the Identity Mapping of the ResNet Architecture.

Date of Submission

December 2020

Date of Award

Winter 12-12-2021

Institute Name (Publisher)

Indian Statistical Institute

Document Type

Master's Dissertation

Degree Name

Master of Technology

Subject Name

Computer Science

Department

Computer Vision and Pattern Recognition Unit (CVPR-Kolkata)

Supervisor

Bhattacharya, Ujjwal (CVPR-Kolkata; ISI)

Abstract (Summary of the Work)

In the past few years, deep models have made a huge impact in the field of computer vision. Among these deep models, Residual Networks or ResNets have become particularly popular for their simple architecture and efficient performance. Despite the achievement, the skip connection which made the training of a very deep model possible was also considered as a drawback of this model. Some studies have been done on the comparative performance of various types of skip connections. Inspired by the recent work on skip connection which proposed use of ReLU with group normalization as an alternative to the identity skip connection resulting in better performance than traditional ResNet, we have explored use of various activation functions. In this thesis, we propose a different transformation to be used together with the ReLU Group Normalization (RG) connection to improve the performance of Residual networks. We simulated our results on CIFAR-10 and CIFAR-100 datasets. The code developed as a part of this study is available at https://github.com/Arpan142/Arpan dissertation.

Comments

ProQuest Collection ID: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:28842743

Control Number

ISI-DISS-2020-07

Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.

DOI

http://dspace.isical.ac.in:8080/jspui/handle/10263/7155

This document is currently not available here.

Share

COinS