CS 335: Fair, Accountable, and Transparent (FAccT) Deep Learning

Final Reports

• Shubhang Desai: Fair Attention-Based Image Captioning.

• Thomas Dimson: Translating from Unfair to Fair Embeddings, recording.

• Eric Frankel and Eddie Vendrow: Fair Generation through Prior Modification, session recording.

• Sasha Harrison and Boxiao Pan: Mitigating Bias in Facial Recognition with FairGAN, session recording.

• Caroline Ho, Hugo Kitano, and Kevin Lee: Fair Image Classification with Semi-Supervised Learning, session recording.

• J. Weston Hughes: Explaining Video Classification and Regression Models, with an Application to Echocardiograms, recording.

• Jacob Reiter: Developing an Interpretable Schizophrenia Deep Learning Classifier on fMRI and sMRI using a Patient-Centered DeepSHAP, session recording.

• Daniel Tang: Interpretability of Deep Learning Models for Classification of Epilepsy in EEG Recordings.

• Chris Waites: Differentially Private Normalizing Flows for Privacy-Preserving Density Estimation, session recording.

• Evani Radiya-Dixit: Protecting Privacy Against Facial Recognition Models.