Vikram Ramaswamy

About

Ramaswamy is interested in fairness and interpretability in machine learning, and how it applies to visual systems. He has worked primarily in constructing better datasets (either real or synthetic) and understanding and evaluating interpretability methods for convolutional neural networks.