Get our free extension to see links to code for papers anywhere online!

 Add to Chrome

 Add to Firefox

CatalyzeX Code Finder - Browser extension linking code for ML papers across the web! | Product Hunt Embed
QuPeL: Quantized Personalization with Applications to Federated Learning

Feb 23, 2021
Kaan Ozkara, Navjot Singh, Deepesh Data, Suhas Diggavi


  Access Paper or Ask Questions

Shuffled Model of Federated Learning: Privacy, Communication and Accuracy Trade-offs

Aug 17, 2020
Antonious M. Girgis, Deepesh Data, Suhas Diggavi, Peter Kairouz, Ananda Theertha Suresh


  Access Paper or Ask Questions

Byzantine-Resilient High-Dimensional SGD with Local Iterations on Heterogeneous Data

Jun 22, 2020
Deepesh Data, Suhas Diggavi

* 31 pages. arXiv admin note: text overlap with arXiv:2005.07866 

  Access Paper or Ask Questions

Successive Refinement of Privacy

May 24, 2020
Antonious M. Girgis, Deepesh Data, Kamalika Chaudhuri, Christina Fragouli, Suhas Diggavi


  Access Paper or Ask Questions

Byzantine-Resilient SGD in High Dimensions on Heterogeneous Data

May 16, 2020
Deepesh Data, Suhas Diggavi

* 57 pages, 2 figures 

  Access Paper or Ask Questions

SQuARM-SGD: Communication-Efficient Momentum SGD for Decentralized Optimization

May 13, 2020
Navjot Singh, Deepesh Data, Jemin George, Suhas Diggavi

* 52 pages, 6 figures. arXiv admin note: substantial text overlap with arXiv:1910.14280 

  Access Paper or Ask Questions

SPARQ-SGD: Event-Triggered and Compressed Communication in Decentralized Stochastic Optimization

Oct 31, 2019
Navjot Singh, Deepesh Data, Jemin George, Suhas Diggavi

* 41 pages, 4 figures 

  Access Paper or Ask Questions

Data Encoding for Byzantine-Resilient Distributed Optimization

Jul 05, 2019
Deepesh Data, Linqi Song, Suhas Diggavi

* 30 pages, 5 figures 

  Access Paper or Ask Questions

Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification, and Local Computations

Jun 06, 2019
Debraj Basu, Deepesh Data, Can Karakus, Suhas Diggavi

* 46 pages, 9 figures 

  Access Paper or Ask Questions