End-to-End Neural Network Compression via l1/l2 Regularized Latency Surrogates.

Anshul Nasery, Hardik Shah, Arun Sai Suggala, Prateek Jain

Browse the full CVPR paper archive.