Stanford
University
  • Stanford Home
  • Maps & Directions
  • Search Stanford
  • Emergency Info
  • Terms of Use
  • Privacy
  • Copyright
  • Trademarks
  • Non-Discrimination
  • Accessibility
© Stanford University.  Stanford, California 94305.
Ehsan Adeli | Stanford HAI
Skip to content
  • About

    • About
    • People
    • Get Involved with HAI
    • Support HAI
    • Subscribe to Email
  • Research

    • Research
    • Fellowship Programs
    • Grants
    • Student Affinity Groups
    • Centers & Labs
    • Research Publications
    • Research Partners
  • Education

    • Education
    • Executive and Professional Education
    • Government and Policymakers
    • K-12
    • Stanford Students
  • Policy

    • Policy
    • Policy Publications
    • Policymaker Education
    • Student Opportunities
  • AI Index

    • AI Index
    • AI Index Report
    • Global Vibrancy Tool
    • People
  • News
  • Events
  • Industry
  • Centers & Labs
Navigate
  • About
  • Events
  • AI Glossary
  • Careers
  • Search
Participate
  • Get Involved
  • Support HAI
  • Contact Us

Stay Up To Date

Get the latest news, advances in research, policy work, and education program updates from HAI in your inbox weekly.

Sign Up For Latest News

peopleFaculty

Ehsan Adeli

Assistant Professor of Psychiatry and Behavioral Sciences, by courtesy, of Biomedical Data Science, and of Computer Science

External Bio
Latest Work
The World Model and Spatial Intelligence Era: Governing AI Beyond Language
Daniel Zhang, Russell Wald, Ehsan Adeli, Elena Cryst, Daniel E. Ho, Caroline Meinhardt, Jiajun Wu, Amy Zegart, Fei-Fei Li
Quick ReadJul 27
issue brief

This brief highlights the emergence of world models and outlines a first-of-its-kind governance and policy agenda for the technology.

Multi-Label, Multi-Domain Learning Identifies Compounding Effects of HIV and Cognitive Impairment
Jiequan Zhang, Qingyu Zhao, Ehsan Adeli, Adolf Pfefferbaum, Edith Sullivan, Robert Paul, Victor Valcour, Kilian Pohl
Mar 20
Research
Your browser does not support the video tag.

Multi-Label, Multi-Domain Learning Identifies Compounding Effects of HIV and Cognitive Impairment

Association of Heavy Drinking With Deviant Fiber Tract Development in Frontal Brain Systems in Adolescents
Qingyu Zhao, Nicolas Honnorat, Ehsan Adeli, Simon Podhajsky, Michael D. De Bellis, James Voyvodic, Kate B. Nooner, Fiona C. Baker, Ian M. Colrain, Susan F. Tapert, Sandra A. Brown, Wesley K. Thompson, Bonnie J. Nagel, Duncan B. Clark
Dec 25
Research
Your browser does not support the video tag.

Association of Heavy Drinking With Deviant Fiber Tract Development in Frontal Brain Systems in Adolescents

Share
Link copied to clipboard!

All Related

Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models
Zixuan Liu, Ehsan Adeli, Kilian M. Pohl, Qingyu Zhao
Dec 18, 2021
Research
Your browser does not support the video tag.

Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models

Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models

Zixuan Liu, Ehsan Adeli, Kilian M. Pohl, Qingyu Zhao
Dec 18, 2021

Going Beyond Saliency Maps: Training Deep Models to Interpret Deep Models

Your browser does not support the video tag.
Research
Longitudinal Correlation Analysis for Decoding Multi-Modal Brain Development
Qingyu Zhao, Ehsan Adeli, Kilian M. Pohl
Dec 14, 2021
Research
Your browser does not support the video tag.

Longitudinal Correlation Analysis for Decoding Multi-Modal Brain Development

Longitudinal Correlation Analysis for Decoding Multi-Modal Brain Development

Qingyu Zhao, Ehsan Adeli, Kilian M. Pohl
Dec 14, 2021

Longitudinal Correlation Analysis for Decoding Multi-Modal Brain Development

Your browser does not support the video tag.
Research
Longitudinal Pooling & Consistency Regularization to Model Disease Progression from MRIs
Jiahong Ouyang, Qingyu Zhao, Adolf Pfefferbaum, Susan F Tapert, Ehsan Adeli, Kilian Pohl
Dec 12, 2021
Research
Your browser does not support the video tag.

Longitudinal Pooling & Consistency Regularization to Model Disease Progression from MRIs

Longitudinal Pooling & Consistency Regularization to Model Disease Progression from MRIs

Jiahong Ouyang, Qingyu Zhao, Adolf Pfefferbaum, Susan F Tapert, Ehsan Adeli, Kilian Pohl
Dec 12, 2021

Longitudinal Pooling & Consistency Regularization to Model Disease Progression from MRIs

Your browser does not support the video tag.
Research
Longitudinal Self-Supervised Learning
Qingyu Zhao, Zixuan Liu, Ehsan Adeli, Kilian M. Pohl
Dec 10, 2021
Research
Your browser does not support the video tag.

Longitudinal Self-Supervised Learning

Longitudinal Self-Supervised Learning

Qingyu Zhao, Zixuan Liu, Ehsan Adeli, Kilian M. Pohl
Dec 10, 2021

Longitudinal Self-Supervised Learning

Your browser does not support the video tag.
Research
MOMA: Multi-Object Multi-Actor Activity Parsing
Zelun Luo, Wanze Xie, Siddharth Kapoor, Yiyun Liang, Michael Cooper, Juan Carlos Niebles, Ehsan Adeli
Dec 09, 2021
Research
Your browser does not support the video tag.

MOMA: Multi-Object Multi-Actor Activity Parsing

MOMA: Multi-Object Multi-Actor Activity Parsing

Zelun Luo, Wanze Xie, Siddharth Kapoor, Yiyun Liang, Michael Cooper, Juan Carlos Niebles, Ehsan Adeli
Dec 09, 2021

MOMA: Multi-Object Multi-Actor Activity Parsing

Your browser does not support the video tag.
Research
Metadata Normalization
Mandy Lu, Qingyu Zhao, Jiequan Zhang, Kilian M. Pohl, Juan Carlos Niebles, Ehsan Adeli
Dec 08, 2021
Research
Your browser does not support the video tag.

Metadata Normalization

Metadata Normalization

Mandy Lu, Qingyu Zhao, Jiequan Zhang, Kilian M. Pohl, Juan Carlos Niebles, Ehsan Adeli
Dec 08, 2021

Metadata Normalization

Your browser does not support the video tag.
Research
Representation Disentanglement for Multi-modal MR Analysis
Jiahong Ouyang, Ehsan Adeli, Kilian M. Pohl, Qingyu Zhao, Greg Zaharchuk
Dec 03, 2021
Research
Your browser does not support the video tag.

Representation Disentanglement for Multi-modal MR Analysis

Representation Disentanglement for Multi-modal MR Analysis

Jiahong Ouyang, Ehsan Adeli, Kilian M. Pohl, Qingyu Zhao, Greg Zaharchuk
Dec 03, 2021

Representation Disentanglement for Multi-modal MR Analysis

Your browser does not support the video tag.
Research
Scalable Differential Privacy with Sparse Network Fine-Tuning
Zelun Luo, Daniel Wu, Ehsan Adeli
Nov 28, 2021
Research
Your browser does not support the video tag.

Scalable Differential Privacy with Sparse Network Fine-Tuning

Scalable Differential Privacy with Sparse Network Fine-Tuning

Zelun Luo, Daniel Wu, Ehsan Adeli
Nov 28, 2021

Scalable Differential Privacy with Sparse Network Fine-Tuning

Your browser does not support the video tag.
Research
Self-Supervised Longitudinal Neighbourhood Embedding
Jiahong Ouyang, Qingyu Zhao, Ehsan Adeli, Adolf Pfefferbaum, Greg Zaharchuk, Kilian Pohl
Nov 27, 2021
Research
Your browser does not support the video tag.

Self-Supervised Longitudinal Neighbourhood Embedding

Self-Supervised Longitudinal Neighbourhood Embedding

Jiahong Ouyang, Qingyu Zhao, Ehsan Adeli, Adolf Pfefferbaum, Greg Zaharchuk, Kilian Pohl
Nov 27, 2021

Self-Supervised Longitudinal Neighbourhood Embedding

Your browser does not support the video tag.
Research
Spatio-Temporal Graph Convolution for Resting-State fMRI Analysis
Soham Gadgil, Qingyu Zhao, Adolf Pfefferbaum, Ehsan Adeli, Kilian M. Pohl
Nov 26, 2021
Research
Your browser does not support the video tag.

Spatio-Temporal Graph Convolution for Resting-State fMRI Analysis

Spatio-Temporal Graph Convolution for Resting-State fMRI Analysis

Soham Gadgil, Qingyu Zhao, Adolf Pfefferbaum, Ehsan Adeli, Kilian M. Pohl
Nov 26, 2021

Spatio-Temporal Graph Convolution for Resting-State fMRI Analysis

Your browser does not support the video tag.
Research
Inpainting Cropped Diffusion MRI using Deep Generative Models
Rafi Ayub, Qingyu Zhao, M.J. Meloy, Edith Sullivan, Adolf Pfefferbaum, Ehsan Adeli, Kilian Pohl
Dec 12, 2020
Research
Your browser does not support the video tag.

Inpainting Cropped Diffusion MRI using Deep Generative Models

Inpainting Cropped Diffusion MRI using Deep Generative Models

Rafi Ayub, Qingyu Zhao, M.J. Meloy, Edith Sullivan, Adolf Pfefferbaum, Ehsan Adeli, Kilian Pohl
Dec 12, 2020

Inpainting Cropped Diffusion MRI using Deep Generative Models

Your browser does not support the video tag.
Research
Representation Learning with Statistical Independence to Mitigate Bias
Ehsan Adeli, Qingyu Zhao, Adolf Pfefferbaum, Edith Sullivan, Fei-Fei Li, Juan Carlos Niebles, Kilian Pohl
Dec 03, 2020
Research

Presence of bias (in datasets or tasks) is inarguably one of the most critical challenges in machine learning applications that has alluded to pivotal debates in recent years. Such challenges range from spurious associations between variables in medical studies to the bias of race in gender or face recognition systems. Controlling for all types of biases in the dataset curation stage is cumbersome and sometimes impossible. The alternative is to use the available data and build models incorporating fair representation learning. In this paper, we propose such a model based on adversarial training with two competing objectives to learn features that have (1) maximum discriminative power with respect to the task and (2) minimal statistical mean dependence with the protected (bias) variable(s). Our approach does so by incorporating a new adversarial loss function that encourages a vanished correlation between the bias and the learned features. We apply our method to synthetic data, medical images (containing task bias), and a dataset for gender classification (containing dataset bias). Our results show that the learned features by our method not only result in superior prediction performance but also are unbiased.

Representation Learning with Statistical Independence to Mitigate Bias

Ehsan Adeli, Qingyu Zhao, Adolf Pfefferbaum, Edith Sullivan, Fei-Fei Li, Juan Carlos Niebles, Kilian Pohl
Dec 03, 2020

Presence of bias (in datasets or tasks) is inarguably one of the most critical challenges in machine learning applications that has alluded to pivotal debates in recent years. Such challenges range from spurious associations between variables in medical studies to the bias of race in gender or face recognition systems. Controlling for all types of biases in the dataset curation stage is cumbersome and sometimes impossible. The alternative is to use the available data and build models incorporating fair representation learning. In this paper, we propose such a model based on adversarial training with two competing objectives to learn features that have (1) maximum discriminative power with respect to the task and (2) minimal statistical mean dependence with the protected (bias) variable(s). Our approach does so by incorporating a new adversarial loss function that encourages a vanished correlation between the bias and the learned features. We apply our method to synthetic data, medical images (containing task bias), and a dataset for gender classification (containing dataset bias). Our results show that the learned features by our method not only result in superior prediction performance but also are unbiased.

Machine Learning
Research
1
2