• About Us
  • Contact Us
  • Terms & Conditions
  • Privacy Policy
Technology Hive
  • Home
  • Technology
  • Artificial Intelligence (AI)
  • Cyber Security
  • Machine Learning
  • More
    • Deep Learning
    • AI in Healthcare
    • AI Regulations & Policies
    • Business
    • Cloud Computing
    • Ethics & Society
No Result
View All Result
  • Home
  • Technology
  • Artificial Intelligence (AI)
  • Cyber Security
  • Machine Learning
  • More
    • Deep Learning
    • AI in Healthcare
    • AI Regulations & Policies
    • Business
    • Cloud Computing
    • Ethics & Society
No Result
View All Result
Technology Hive
No Result
View All Result
Home Artificial Intelligence (AI)

Researchers reduce bias in AI models while preserving or improving accuracy

Adam Smith – Tech Writer & Blogger by Adam Smith – Tech Writer & Blogger
March 7, 2025
in Artificial Intelligence (AI)
0
Researchers reduce bias in AI models while preserving or improving accuracy
0
SHARES
0
VIEWS
Share on FacebookShare on Twitter

Machine Learning Models Can Fail When They Try to Make Predictions for Individuals Who Were Underrepresented in the Datasets They Were Trained On

Removing Bad Examples

Machine learning models can fail when they try to make predictions for individuals who were underrepresented in the datasets they were trained on. For instance, a model that predicts the best treatment option for someone with a chronic disease may be trained using a dataset that contains mostly male patients. That model might make incorrect predictions for female patients when deployed in a hospital.

A New Technique

To improve outcomes, engineers can try balancing the training dataset by removing data points until all subgroups are represented equally. While dataset balancing is promising, it often requires removing a large amount of data, hurting the model’s overall performance.

MIT researchers have developed a new technique that identifies and removes specific points in a training dataset that contribute most to a model’s failures on minority subgroups. By removing far fewer data points than other approaches, this technique maintains the overall accuracy of the model while improving its performance regarding underrepresented groups.

How it Works

The researchers combined two ideas: the idea that machine learning models are trained using huge datasets gathered from many sources across the internet, and the idea that some data points impact a model’s performance on certain downstream tasks more than others. They developed an approach that identifies and removes these problematic datapoints.

The researchers take incorrect predictions the model made about minority subgroups and use a method called TRAK to identify which training examples contributed the most to that incorrect prediction. By aggregating this information across bad test predictions in the right way, they are able to find the specific parts of the training that are driving worst-group accuracy down overall.

Results

The researchers tested their technique on three machine-learning datasets and found that it outperformed multiple techniques. In one instance, it boosted worst-group accuracy while removing about 20,000 fewer training samples than a conventional data balancing method. Their technique also achieved higher accuracy than methods that require making changes to the inner workings of a model.

A More Accessible Approach

Because the MIT method involves changing a dataset instead of a model, it would be easier for a practitioner to use and can be applied to many types of models. It can also be used when bias is unknown because subgroups in a training dataset are not labeled. By identifying datapoints that contribute most to a feature the model is learning, they can understand the variables it is using to make a prediction.

Conclusion

The researchers’ new technique is a tool that can be used to detect and remove sources of bias in machine learning models. By identifying and removing specific points in a training dataset that contribute to a model’s failures on minority subgroups, this technique can improve the performance of machine learning models on underrepresented groups.

Frequently Asked Questions

Q: How does the new technique work?
A: The technique identifies and removes specific points in a training dataset that contribute most to a model’s failures on minority subgroups.

Q: What are the benefits of this technique?
A: The technique maintains the overall accuracy of the model while improving its performance on minority subgroups, and can be used when bias is unknown because subgroups in a training dataset are not labeled.

Q: How does it compare to other techniques?
A: The technique outperforms multiple techniques, including dataset balancing and making changes to the inner workings of a model.

Previous Post

CMU Research Shows Compression Alone May Unlock AI Puzzle-Solving Abilities

Next Post

Information for Busy People

Adam Smith – Tech Writer & Blogger

Adam Smith – Tech Writer & Blogger

Adam Smith is a passionate technology writer with a keen interest in emerging trends, gadgets, and software innovations. With over five years of experience in tech journalism, he has contributed insightful articles to leading tech blogs and online publications. His expertise covers a wide range of topics, including artificial intelligence, cybersecurity, mobile technology, and the latest advancements in consumer electronics. Adam excels in breaking down complex technical concepts into engaging and easy-to-understand content for a diverse audience. Beyond writing, he enjoys testing new gadgets, reviewing software, and staying up to date with the ever-evolving tech industry. His goal is to inform and inspire readers with in-depth analysis and practical insights into the digital world.

Related Posts

UK and Singapore Form AI Finance Alliance
Artificial Intelligence (AI)

UK and Singapore Form AI Finance Alliance

by Adam Smith – Tech Writer & Blogger
July 4, 2025
CyXcel Research Uncovers AI Risks for UK Businesses
Artificial Intelligence (AI)

CyXcel Research Uncovers AI Risks for UK Businesses

by Adam Smith – Tech Writer & Blogger
July 3, 2025
Don’t Let Hype Exceed Reality on AI Agents
Artificial Intelligence (AI)

Don’t Let Hype Exceed Reality on AI Agents

by Adam Smith – Tech Writer & Blogger
July 3, 2025
The AI Energy Paradox
Artificial Intelligence (AI)

The AI Energy Paradox

by Adam Smith – Tech Writer & Blogger
July 2, 2025
AI Can Slash Global Carbon Emissions
Artificial Intelligence (AI)

AI Can Slash Global Carbon Emissions

by Adam Smith – Tech Writer & Blogger
July 2, 2025
Next Post
Information for Busy People

Information for Busy People

Leave a Reply Cancel reply

Your email address will not be published. Required fields are marked *

Latest Articles

Gemini to Replace Google Assistant Later This Year

Gemini to Replace Google Assistant Later This Year

March 15, 2025
InterSystems at HIMSS25: Laser Focused on Helping Clients with GenAI

InterSystems at HIMSS25: Laser Focused on Helping Clients with GenAI

March 3, 2025
To D-ID or not to D-ID?

To D-ID or not to D-ID?

March 5, 2025

Browse by Category

  • AI in Healthcare
  • AI Regulations & Policies
  • Artificial Intelligence (AI)
  • Business
  • Cloud Computing
  • Cyber Security
  • Deep Learning
  • Ethics & Society
  • Machine Learning
  • Technology
Technology Hive

Welcome to Technology Hive, your go-to source for the latest insights, trends, and innovations in technology and artificial intelligence. We are a dynamic digital magazine dedicated to exploring the ever-evolving landscape of AI, emerging technologies, and their impact on industries and everyday life.

Categories

  • AI in Healthcare
  • AI Regulations & Policies
  • Artificial Intelligence (AI)
  • Business
  • Cloud Computing
  • Cyber Security
  • Deep Learning
  • Ethics & Society
  • Machine Learning
  • Technology

Recent Posts

  • Agentic AI Replaces White-Collar Workflows
  • Building Intelligent Workflows with AI Tools
  • Optimize Machine Learning Models with Hyperparameter Tuning
  • Corrective Retrieval-Augmented Generation Model
  • Will AI Replace Humans?

Our Newsletter

Subscribe Us To Receive Our Latest News Directly In Your Inbox!

We don’t spam! Read our privacy policy for more info.

Check your inbox or spam folder to confirm your subscription.

© Copyright 2025. All Right Reserved By Technology Hive.

No Result
View All Result
  • Home
  • Technology
  • Artificial Intelligence (AI)
  • Cyber Security
  • Machine Learning
  • AI in Healthcare
  • AI Regulations & Policies
  • Business
  • Cloud Computing
  • Ethics & Society
  • Deep Learning

© Copyright 2025. All Right Reserved By Technology Hive.

Are you sure want to unlock this post?
Unlock left : 0
Are you sure want to cancel subscription?