Publications

Not Too Close and Not Too Far Enforcing Monotonicity Requires Penalizing The Right Points

Optimization; Responsible AI

Max-Margin Adversarial Training: Direct Input Space Margin Maximization through Adversarial Training

Learning And Generalization; Responsible AI

Adapting Grad-CAM for Embedding Networks

Computer Vision; Responsible AI

Privacy-Preserving Q-Learning with Functional Noise in Continuous Spaces

Responsible AI

On the Effectiveness of Low Frequency Perturbations

Computer Vision; Responsible AI

Private Q-Learning with Functional Noise in Continuous Spaces

Reinforcement Learning; Responsible AI

On the Sensitivity of Adversarial Robustness to Input Data Distributions

Learning And Generalization; Responsible AI

Compositional Hard Negatives for Visual Semantic Embeddings via an Adversary

Computer Vision; Natural Language Processing; Responsible AI

Adversarial Contrastive Estimation

Learning And Generalization; Responsible AI

Improving GAN Training via Binarized Representation Entropy (BRE) Regularization

Learning And Generalization; Responsible AI