Gender Biases in Computer Vision Models

About the video

Abhishek Mandal, a PhD researcher whose focus is computer vision (and one of Women at the Table’s first Tech Fellows) focusses on Generated Bias: Gender and Geographic Bias in Generative AI.

His research has been groundbreaking in terms of focus on both gender and geography in multi-modal contexts.

This is a simple primer on his work based on a recent paper Measuring Bias in Multimodal Models: Multimodal Composite Association Score  (which seeks to help mitigate, or at least score, the problem).

About the author

Abhishek Mandal is a SFI funded PhD researcher at the Insight Centre for Data Analytics, under the supervision of Dr. Suzanne Little from Dublin City University and Dr. Susan Leavy from University College Dublin. His research centers on auditing bias and imbalance within deep neural networks employed in computer vision applications. Abhishek’s areas of expertise encompass Convolutional Neural Networks, Generative Adversarial Networks, Deep Learning, Computer Vision, Multimedia Analytics, and the examination of Ethics and Bias in AI.

Recommended resources

→ Mandal, A., Leavy, S. and Little, S., 2023, April. Measuring Bias in Multimodal Models: Multimodal Composite Association Score. In International Workshop on Algorithmic Bias in Search and Recommendation (pp. 17-30). Cham: Springer Nature Switzerland.


Join our community

We are committed to advancing human rights-based approaches in AI & encourage anyone interested in learning more from a global perspective to explore and contribute to our community!