Communication Efficient Distributed Learning (Springer)
Authors: Navjot Singh, Deepesh Data, Suhas Diggavi
Book Title: Artificial Intelligence for Edge Computing (Aug. 2023)
Editors: Mudhakar Srivatsa, Tarek Abdelzaher, Ting He
Decentralized Optimization Resilient Against Local Data Poisoning Attacks (IEEEXplore)
Authors: Yanwen Mao, Deepesh Data, Suhas Diggavi, Paulo Tabuada
IEEE Transactions on Automatic Control, vol. 70, no. 1, pp. 81-96, Jan. 2025.
Byzantine-Resilient High-Dimensional Federated Learning (IEEEXplore)
Authors: Deepesh Data, Suhas Diggavi
IEEE Transactions on Information Theory, vol. 69, no. 10, pp. 6639-6670, Oct. 2023.
Must the Communication Graph of MPC Protocols be an Expander? (Full Version)
Authors: Elette Boyle, Ran Cohen, Deepesh Data, Pavel Hubacek
Journal of Cryptology, vol. 36, no. 3, May 2023.
SPARQ-SGD: Event-Triggered and Compressed Communication in Decentralized Optimization (IEEEXplore, arXiv)
Authors: Navjot Singh, Deepesh Data, Jemin George, Suhas Diggavi
IEEE Transactions on Automatic Control, vol. 68, no. 2, pp. 721-736, Feb. 2023.
SQuARM-SGD: Communication-Efficient Momentum SGD for Decentralized Optimization (IEEEXplore, arXiv)
Authors: Navjot singh, Deepesh Data, Jemin George, Suhas Diggavi
IEEE Journal on Selected Areas in Information Theory, vol. 2, no. 3, pp. 954-969, Sept. 2021.
Shuffled Model of Federated Learning: Privacy, Communication, and Accuracy Trade-offs (IEEEXplore, arXiv)
Authors: Antonious Girgis, Deepesh Data, Suhas Diggavi, Peter Kairouz, Ananda Theertha Suresh
IEEE Journal on Selected Areas in Information Theory, vol. 2, no. 1, pp. 464-478, March 2021.
Data Encoding for Byzantine-Resilient Distributed Optimization (IEEEXplore, arXiv)
Authors: Deepesh Data, Linqi Song, Suhas Diggavi
IEEE Transactions on Information Theory, vol. 67, no. 2, pp. 1117-1140, Feb. 2021.
Successive Refinement of Privacy (IEEEXplore, arXiv)
Authors: Antonious Girgis, Deepesh Data, Kamalika Choudhuri, Christina Fragouli, Suhas Diggavi
IEEE Journal on Selected Areas in Information Theory, vol. 1, no. 3, pp. 745-759, Nov. 2020.
Interactive Secure Function Computation (IEEEXplore, arXiv)
Authors: Deepesh Data, Gowtham R. Kurri, Jithin Ravi, Vinod M. Prabhakaran
IEEE Transactions on Information Theory, vol. 66, no. 9, pp. 5492 - 5521, Sept. 2020.
Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification, and Local Computations (IEEEXplore, arXiv, Slides)
Authors: Debraj Basu, Deepesh Data, Can Karakus, Suhas Diggavi
IEEE Journal on Selected Areas in Information Theory, vol. 1, no. 1, pp. 217-226, May 2020.
Communication and Randomness Lower Bounds for Secure Computation (IEEEXplore, arXiv)
Authors: Deepesh Data, Vinod M. Prabhakaran, Manoj Prabhakaran
IEEE Transactions on Information Theory, vol. 62, no. 7, pp. 3901-3929, July 2016
Utilitarian Privacy and Private Sampling (IEEEXplore)
Authors: Aman Bansal, Rahul Chunduru, Deepesh Data, Manoj Prabhakaran
IEEE International Symposium on Information Theory (ISIT), 2024, Athens, Greece.
A Generative Framework for Personalized Learning and Estimation: Theory, Algorithms, and Privacy (arXiv)
Authors: Kaan Ozkara, Antonious Girgis, Deepesh Data, Suhas Diggavi
International Conference on Learning Representation (ICLR) 2023, Kigali, Rwanda.
Decentralized Learning Robust to Data Poisoning Attacks (IEEEXplore)
Authors: Yanwen Mao, Deepesh Data, Suhas Diggavi, Paulo Tabuada
IEEE Conference on Decision and Control (CDC), 2022, Cancún, Mexico.
Distributed User-Level Private Mean Estimation (IEEEXplore)
Authors: Antonious Girgis, Deepesh Data, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2022, Finland.
Flexible Accuracy for Differential Privacy (PMLR, arXiv)
Authors: Aman Bansal, Rahul Chunduru, Deepesh Data, Manoj Prabhakaran
International Conference on Artificial Intelligence and Statistics (AISTATS) 2022.
Theory and Practice of Differential Privacy (TPDP), 2021, an ICML workshop (link)
Renyi Differential Privacy of the Subsampled Shuffle Model in Distributed Learning (arXiv)
Authors: Antonious Girgis, Deepesh Data, Suhas Diggavi
Neural Information Processing Systems (NeurIPS), 2021, (virtual).
QuPeD: Quantized Personalization via Distillation with Applications to Federated Learning (arXiv)
Authors: Kaan Ozkara, Navjot Singh, Deepesh Data, Suhas Diggavi
Neural Information Processing Systems (NeurIPS), 2021, (virtual).
On the Renyi Differential Privacy of the Shuffle Model (arXiv)
Authors: Antonious Girgis, Deepesh Data, Suhas Diggavi, Ananda Theertha Suresh, Peter Kairouz
ACM Conference on Computer and Communications Security (CCS) 2021, (virtual) -- Best Paper Award
Theory and Practice of Differential Privacy (TPDP), 2021, an ICML workshop (link)
Byzantine-Resilient High-Dimensional SGD with Local Iterations on Heterogeneous Data (PMLR, arXiv)
Authors: Deepesh Data, Suhas Diggavi
International Conference on Machine Learning (ICML), 2021, (virtual).
Shuffled Model of Differential Privacy in Federated Learning (PMLR, arXiv)
Authors: Antonious Girgis, Deepesh Data, Suhas Diggavi, Peter Kairouz, Ananda Theertha Suresh
International Conference on Artificial Intelligence and Statistics (AISTATS) 2021 (virtual).
NeurIPS workshop on privacy-preserving machine learning (PPML) 2020. (link) -- Oral Presentation
Byzantine-Resilient SGD in High Dimensions on Heterogeneous Data (arXiv)
Authors: Deepesh Data, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2021, (virtual).
Differential Private Federated Learning with Shuffling and Client Self-Sampling
Authors: Antonious Girgis, Deepesh Data, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2021, (virtual).
SQuARM-SGD: Communication-Efficient Momentum SGD for Decentralized Optimization (arXiv)
Authors: Navjot Singh, Deepesh Data, Jemin George, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2021, (virtual).
SPARQ-SGD: Event-Triggered and Compressed Communication in Decentralized Optimization (arXiv)
Authors: Navjot Singh, Deepesh Data, Jemin George, Suhas Diggavi
IEEE Conference on Decision and Control (CDC), 2020, Jeju Island, Republic of Korea.
Hiding Identities: Estimation Under Local Differential Privacy
Authors: Antonious M. Girgis, Deepesh Data, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2020, (virtual).
On Byzantine-Resilient High-Dimensional Stochastic Gradient Descent
Authors: Deepesh Data, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2020, (virtual).
Qsparse-local-SGD: Distributed SGD with Quantization, Sparsification, and Local Computations (arXiv, NeurIPS, Slides)
Authors: Debraj Basu, Deepesh Data, Can Karakus, Suhas Diggavi
Neural Information Processing Systems (NeurIPS), 2019, Vancouver, Canada.
Byzantine-Tolerant Distributed Coordinate Descent
Authors: Deepesh Data, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2019, Paris, France.
Data Encoding Methods For Byzantine-Resilient Distributed Optimization
Authors: Deepesh Data, Linqi Song, Suhas Diggavi
IEEE International Symposium on Information Theory (ISIT), 2019, Paris, France.
Data Encoding For Byzantine-Resilient Distributed Gradient Descent (Invited Paper) (IEEEXplore)
Authors: Deepesh Data, Linqi Song, Suhas Diggavi
Allerton Conference on Communication, Control, and Computing, 2018, Monticello, Illinois, USA.
Must the Communication Graph of MPC Protocols be an Expander? (Springer, Full Version)
Authors: Elette Boyle, Ran Cohen, Deepesh Data, Pavel Hubacek
Advances in Cryptology - CRYPTO, 2018, Santa Barbara, California, USA.
Towards Characterizing Securely Computable Two-Party Randomized Functions (Springer, ePrint)
Authors: Deepesh Data, Manoj Prabhakaran
Public Key Cryptography (PKC), 2018, Rio De Janeiro, Brazil.
Secure Computation of Randomized Functions: Further Results (IEEEXplore, arXiv)
Authors: Deepesh Data, Vinod M. Prabhakaran
IEEE Information Theory Workshop (ITW), 2017, Kaohsiung, Taiwan.
Secure Computation of Randomized Functions (IEEEXplore, arXiv)
Author: Deepesh Data
IEEE International Symposium on Information Theory (ISIT), 2016, Barcelona, Spain.
On Coding for Secure Computing (IEEEXplore, Full version)
Authors: Deepesh Data, Vinod M. Prabhakaran
IEEE International Symposium on Information Theory (ISIT), 2015, Hong Kong.
How to Securely Compute the Modulo-Two Sum of Binary Sources (IEEEXplore, arXiv)
Authors: Deepesh Data, Vinod M. Prabhakaran, Manoj Mishra, Bikash Kumar Dey
IEEE Information Theory Workshop (ITW), 2014, Hobart, Tasmania, Australia.
On the Communication Complexity of Secure Computation (Springer, Extended arXiv)
Authors: Deepesh Data, Manoj Prabhakaran, Vinod M. Prabhakaran
Advances in Cryptology - CRYPTO, 2014, Santa Barbara, California, USA.
Communication Requirements for Secure Computation (Invited Paper) (IEEEXplore)
Authors: Deepesh Data, Vinod M. Prabhakaran
Allerton Conference on Communication, Control, and Computing, 2013, Monticello, Illinois, USA.
Communication Complexity and Characterization Results in Secure Computation (PDF)
Winner of the ACM India Doctoral Dissertation Award for 2017-18 (Honourable Mention)
TIFR-Sasken Best Ph.D. Thesis Award in Technology and Computer Sciences for 2017-18
Author: Deepesh Data
Advisor: Prof. Vinod M. Prabhakaran
Year: 2017 (final version submitted in January 2018)
Tata Institute of Fundamental Research, Mumbai, India