Sathya N. Ravi

Optimizing Nondecomposable Data Dependent Regularizers via Lagrangian Reparameterization offers Significant Performance and Efficiency Gains

Sathya N. Ravi, Abhay Venkatesh, Glenn Moo Fung, 2019, AAAI.

Adaptive Activation Thresholding: Dynamic Routing Type Behavior for Interpretability in Convolutional Neural Networks

Sathya N. Ravi, Yiyou Sun, Vikas Singh, 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

Explicitly Imposing Constraints in Deep Networks via Conditional Gradients Gives Improved Generalization and Faster Convergence

Sathya N. Ravi, Vikas Singh, Tuan Dinh, 2019, AAAI.

Physarum Powered Differentiable Linear Programming Layers and Applications pdf

Sathya N. Ravi, Vikas Singh, Zihang Meng, 2020, ArXiv.

FairALM: Augmented Lagrangian Method for Training Fair Models with Little Regret

Sathya N. Ravi, Vikas Singh, Aditya Kumar Akash, 2020, ECCV.

Generating Accurate Pseudo-Labels in Semi-Supervised Learning and Avoiding Overconfident Predictions via Hermite Polynomial Activations

Sathya N. Ravi, Abhay Venkatesh, Vikas Singh, 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).