Skip to Content

This browser is no longer supported.

For the best WWT.com experience, please use one of our supported browsers.

Safari Logo Safari Chrome Logo Chrome Firefox Logo Firefox Edge Logo Edge
Search wwt.com...
Top page results

See all search results

Featured Solutions
What's trending
Help Center
Home
Solutions & Services
Solutions
AI and Data
Automation & Orchestration
Cloud
Data Center
Digital
Digital Workspace
ESG
Mobility
Networking
Security Transformation
See all Solutions
See all Solutions
Services
Application Services
ATC Lab Services
Consulting Services
Customer Success
Infrastructure Services
Mergers & Acquisitions
Strategic Resourcing
Supply Chain & Integration
See all Services
See all Services
Industries
Utilities
Financial Services
Global Service Provider
Healthcare
Life Sciences
Manufacturing
Oil & Gas
Public Sector
Retail
See all Industries
See all Industries
Partners
Cisco
Dell Technologies
HPE
NetApp
VMware
f5
Intel
Microsoft
Palo Alto
See all Partners
See all Partners
Learning & Support
ATC
Communities
Events
Labs
Learning Paths
Research
About
Footer Links
Blog
Careers
Contact Us
Diversity & Inclusion
Locations
News
Sustainability
AI Solutions ATC White Paper Artificial Intelligence and Data
WWT Research • Applied Research Report
• November 9, 2021 • 23 minute read

Mitigating Bias in AI Using Debias-GAN

In this white paper, we propose a general framework, debias-GAN, to address possible bias in AI and Machine Learning (ML) algorithms by explicitly augmenting a training dataset for NLP models with underrepresented instances synthesized by a pretrained sequence generating model.

Abstract

Today's AI and Machine Learning (ML) algorithms have achieved spectacular results in automating decisions that were traditionally made by humans. However, the actual data used for model training may be imbalanced and may introduce discriminatory biases towards specific groups of people. Natural Language Processing (NLP) machine learning models are gaining popularity in various contexts such as resume screening, college admission, emotion assessment, repeated crime prediction, and more. Consequently, it becomes increasingly important to recognize the role they play in contributing to societal biases and stereotypes. NLP models trained on historical data often lack optimization for reducing implicit biases, and in some cases, they further perpetuate biases. Bias in machine learning models presents itself as a strong association amongst attributes that ought not be correlated. In this white paper, we propose a general framework, debias-GAN, to address this issue by explicitly augmenting a training dataset for NLP models with underrepresented instances synthesized by a pretrained sequence generating model. As a proof-of-concept, we chose to experiment with a deep classification model that mimics decorrelation between user ethnicity and tweets. The synthetic data is generated by a targeted language model (LM) that generates realistic but user-ethnicity-oblivious tweets. We trained such debiased LMs with generative adversarial networks (GAN) through reinforcement learning (RL) by adding a penalty function term to the loss function, to minimize sequences with strong indication of user ethnicity via a policy update. The reward is provided by an independently trained classifier that identifies user ethnicity from tweets. We experimented with the ratio of mixed datasets and tested the debiasing impact using three fairness metrics. The debias-GAN is able to improve the fairness metrics of the classifier by up to seven times while maintaining classification performance.  

"WWT Research reports provide in-depth analysis of the latest technology and industry trends, solution comparisons and expert guidance for maturing your organization's capabilities. By logging in or creating a free account you’ll gain access to other reports as well as labs, events and other valuable content."

Thanks for reading. Want to continue?

Log in or create a free account to continue viewing Mitigating Bias in AI Using Debias-GAN and access other valuable content.

© World Wide Technology. All Rights Reserved
  • About
  • Blog
  • Careers
  • Locations
  • News
  • Press Kit
  • Contact Us
  • Privacy Policy
  • Acceptable Use Policy
  • Quality
  • Information Security
  • Supplier Management
  • Cookies