You are here

Machine Learning based Domain Adaptation (MLB-DA) for Multiple Source Classification and Fusion

Award Information
Agency: Department of Defense
Branch: Air Force
Contract: FA8649-20-P-0350
Agency Tracking Number: F19C-002-0077
Amount: $150,000.00
Phase: Phase I
Program: STTR
Solicitation Topic Code: AF19C-T002
Solicitation Number: 19.C
Solicitation Year: 2019
Award Year: 2020
Award Start Date (Proposal Award Date): 2019-12-12
Award End Date (Contract End Date): 2020-12-12
Small Business Information
20271 Goldenrod Lane Suite 2066
Germantown, MD 20876
United States
DUNS: 967349668
HUBZone Owned: No
Woman Owned: Yes
Socially and Economically Disadvantaged: No
Principal Investigator
 Genshe Chen
 (240) 481-5397
Business Contact
 Yingli Wu
Phone: (949) 596-0057
Research Institution
 The University of Memphis
 Stephanie A. Thompson Stephanie A. Thompson
315 Administration Building Division of Research and Innovation
Memphis, TN 38152
United States

 (901) 678-4146
 Nonprofit College or University

Generalizing models learned on one domain to novel domains has been a major obstacle in the quest for universe object recognition. The performance of the learned models degrades significantly when testing on novel domains due to the presence of domain shift. In this proposal, we aim to develop a deep learning-based multi-source self-correcting approach to fuse data with different modalities at the data-level to maximize their capabilities to detect unanticipated events/targets, by leveraging our previous experience in machine learning and heterogeneous data fusion. In this proposal, we propose a Machine Learning based Domain Adaptation (MLB-DA) that leverages unsupervised data to bridge the source and target domain distributions closer in a learned joint feature space. The proposed deep neural network approach holds great capability of adapting to changes of the input distribution allowing self-correcting multiple source classification and fusion. It is focused on learning features that combine (i) discriminativeness and (ii) domain-invariance. The classifiers can adapt to the target domain with different distribution without retraining new input data.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government