An optimal end-to-end training strategy for multi-expert-based neural architecture

Hyungtae Lee; Heesung Kwon

doi:10.1117/12.2558553

21 April 2020 An optimal end-to-end training strategy for multi-expert-based neural architecture

Hyungtae Lee, Heesung Kwon

Proceedings Volume 11413, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications II; 114130L (2020) https://doi.org/10.1117/12.2558553
Event: SPIE Defense + Commercial Sensing, 2020, Online Only

Abstract

Recently, we introduced a state-of-the-art object detection approach referred to as Multi-Expert R-CNN (ME R-CNN) that featured multiple expert classifiers, each being responsible for recognizing objects with distinctive geometrical features. The ME R-CNN architecture consists of multiple components: a shared convolutional network, Multi-Expert classifiers (ME), and Expert Assignment Network (EAN). Both ME and EAN take as a common input the output from the convolutional network and also use each other's output during training. Thus, it is quite challenging to properly train all the components simultaneously to globally optimize the network parameters. The main innovation of the proposed work is to optimize the entire architecture by using a novel training strategy in which manually associated 'RoI-to-expert' mapping is used instead of using the direct output of ME for training EAN. Our experiments show that the proposed training strategy speeds up training time at least 4.2x while maintaining comparable object detection accuracy.

Conference Presentation

Citation Download Citation

Hyungtae Lee and Heesung Kwon "An optimal end-to-end training strategy for multi-expert-based neural architecture", Proc. SPIE 11413, Artificial Intelligence and Machine Learning for Multi-Domain Operations Applications II, 114130L (21 April 2020); https://doi.org/10.1117/12.2558553

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available