Publications (Google Scholar)

Thesis

• Inference and Optimization over Networks: Communication Efficiency and Optimality, Department of Electrical and Computer Engineering, Carnegie Mellon University, 2018

Book Chapter

• A.K. Sahu and S. Kar, Distributed recursive testing of composite hypothesis in multi-agent networks, in Data Fusion in Wireless Sensor Networks: A Statistical Signal Processing Perspective, Domenico Ciuonzo and Pierluigi Salvo Rossi, Eds., Institution of Engineering & Technology, 2019

Patents

• Improved Adversarial Training Using Meta-Learned Initialization, X. Zhang, A. K. Sahu, Z. Kolter, US Patent App. No. 17/062385

• Multiplicative Filter Networks, D. Willmott, A. K. Sahu, R. Fathony, F. Condessa, Z. Kolter, US Patent App. No. 17/034496

• System and Method of a Monotone Operator Neural Network, E. Winston, Z. Kolter, A. K. Sahu, US Patent App. No. 16/850816

• Bayesian Optimization based query efficient black box adversarial attacks, S. N. Shukla, A. K. Sahu, D. Willmott, Z. Kolter, US Patent App. No. 16/580587

• Efficient Black Box adversarial attacks exploiting input data structure, A. K. Sahu, Z. Kolter, US Patent App. No. 16/580650

• Training a machine learning model using a batch based active learning approach, J. Szurley, W. Lin, A. K. Sahu, G. Gupta, US Patent App. No. 16/582928

Preprints

• A. S. Bedi, C. Fan, A. K. Sahu, A. Koppel, F. Huang, B. Saddler and D. Manocha, FedBC: Calibrating Global and Local Models via Federated Learning Beyond Consensus, May 2022

• D, Willmott*, A.K. Sahu*, F. Sheikholeslami*, F. Condessa, and J.Zico Kolter, You only query once: Effective black box adversarial attacks with minimal repeated queries, June 2020 (* denotes equal contribution)

• S.N. Shukla, A.K. Sahu, D. Willmott and J.Zico Kolter, Hard Label Black-box Adversarial Attacks in Low Query Budget Regimes, February 2020

• A.K. Sahu, J.Zico Kolter and S.N. Shukla, Gaussian MRF Covariance Modeling for Efficient Black-Box Adversarial Attacks, September 2019, (Code)

• S.N. Shukla, A.K. Sahu, D. Willmott and J.Zico Kolter, Black-Box Adversarial Attacks with Bayesian Optimization, September 2019

• G. Gupta, A.K. Sahu and Wan-Yi Lin, Learning in Confusion: Batch Active Learning with Noisy Oracle, September 2019

• A.K. Sahu, D.Jakovetic, D. Bajovic and S. Kar, Communication-Efficient Distributed Strongly Convex Stochastic Optimization: Non-Asymptotic Rates

• A.K. Sahu, D.Jakovetic and S. Kar, “Communication Optimality Trade-offs For Distributed Estimation”

Journal Papers

• S. Lee, A.K. Sahu, C. He, S, Avestimehr, Partial model averaging in federated learning: Performance guarantees and benefits, NeuroComputing, 2023

• D. Jakovetic, M. Vukovic, D. Bajovic, A.K. Sahu, S. Kar, Distributed Recursive Estimation under Heavy-Tail Communication Noise, SIAM Journal on Control and Optimization 2023

• D. Jakovetic, D. Bajovic, A.K. Sahu, S. Kar, N. Milosevic and D. Stamenkovic, Nonlinear gradient mappings and stochastic optimization: A general framework with applications to heavy-tail noise, SIAM Journal of Optimization

• N. Petrovic, D. Bajovic, S. Kar, D. Jakovetic, A.K. Sahu, Large deviations for products of non identically distributed network matrices with applications to communication-efficient distributed learning and inference, IEEE Transactions on Signal Processing

• A.K. Sahu and S. Kar, Decentralized Zeroth Order Constrained Stochastic Optimization Algorithms: Frank-Wolfe and Variants With Applications to Black-Box Adversarial Attacks, In the Proceedings of the IEEE.

• T. Li, A.K. Sahu, A. Talwalkar and V. Smith, Federated Learning: Challenges, Methods and Future Directions, August 2019, A shorter version will appear in the Special Issue on Distributed, Streaming Machine Learning of IEEE Signal Processing Magazine 2020.

• J. Wang, A.K. Sahu, G. Joshi and S. Kar, MATCHA: Speeding up Decentralized SGD via Matching Decomposition Sampling, May 2019. Accepted. To appear in IEEE Transactions on Signal Processing. (An abridged version appears in NeurIPS 2019 Workshop on Federated Learning Best Student Paper Award)

• A.K. Sahu, D.Jakovetic, D. Bajovic and S. Kar, Communication Efficient Distributed Weighted Non-Linear Least Squares Estimation, EURASIP Journal on Advances in Signal Processing.Vol.66, No.1, October 2018

• A.K. Sahu, D.Jakovetic and S. Kar, “CIRFE: A Distributed Random Fields Estimator”, IEEE Transactions on Signal Processing, Vol.66, No. 18, pp. 4980-4995, September 2018

• A.K. Sahu and S. Kar, "Recursive Distributed Detection for Composite Hypothesis Testing: Nonlinear Observation Models in Additive Gaussian Noise”, IEEE Transactions on Information Theory. Vol: 63, Issue 8, pp. 4797-4828, 2017.

• A.K. Sahu, S. Kar, Jose’ M. F. Moura, H. Vincent Poor, “Distributed Constrained Recursive Nonlinear Least-Squares Estimation: Algorithms and Asymptotics”, Special Issue on Inference and Learning over Networks of IEEE Transactions on Signal and Information Processing over Networks. Vol:2, Issue 4, pp. 426-441, 2016.

• A.K. Sahu and S. Kar, “Distributed Sequential Detection for Gaussian Shift-in-Mean Hypothesis Testing”, IEEE Transactions on Signal Processing, Vol:64, Issue 1, pp. 89-103, 2016

Conference Papers

• F. Kang, H. A. Just, A. K. Sahu, R. Jia, Performance Scaling via Optimal Transport: Enabling Data Selection from Partially Revealed Sources, NeurIPS 2023

• G.V. Rames, G. Chennupati, M. Rao, A.K. Sahu, A. Rastrow, J. Droppo,Federated Representation Learning for Automatic Speech Recognition, InterSpeech 2023

• A. Agrawal, M. Rao, A.K. Sahu, G. Chennupati, A. Stolcke, Learning when to trust which teacher for weakly supervised ASR, InterSpeech 2023

• M. Rao, G. Chennupati, G. Tiwari, A.K. Sahu, A .Raju, A. Rastrow, J. Droppo, Federated self-learning with weak supervision for speech recognition, ICASSP 2023

• H. Chen, J. Ding, E. Tramel, S. Wu, A.K. Sahu, S. Avestimehr and T. Zhang, Self-Aware Personalized Federated Learning, NeurIPS 2022

• G. Chennupati, M. Rao, G. Chadha, A. Eakin, R. Anirudh, G. Tiwari, A.K. Sahu, A. Rastrow, J. Droppo, A. Oberlin, N. Buddha, P. Venkataraman, Z. Wu and P. Sitpure ,ILASR: Privacy-Preserving Incremental Learning for Automatic Speech Recognition at Production Scale, KDD 2022

• J. Ding, E. Tramel, S. Wu, A.K. Sahu, S. Avestimehr and T. Zhang, Federated learning challenges and opportunities: An outlook, ICASSP 2022

• H. Chen, J. Ding, E. Tramel, S. Wu, A.K. Sahu, S. Avestimehr and T. Zhang, ActPerFL: Active personalized federated learning, ACL 2022 Workshop on Federated Learning for Natural Language Processing(Best Student Paper Award)

• S.Lee, A.K. Sahu, C. He and S. Avestimehr, Partial Model Averaging in Federated Learning: Performance Guarantees and Benefits, AAAI 2022 Workshop on Trustable, Verifiable and Auditable Federated Learning (FL-AAAI-22)

• S.N. Shukla, A.K. Sahu, D. Willmott and J. Zico Kolter, Simple and Efficient Hard Label Black-box Adversarial Attacks in Low Query Budget Regimes, KDD 2021

• R. Fathony*, A.K. Sahu*, D. Willmott* and J. Zico Kolter, Multiplicative Filter Networks, ICLR 2021 (* denotes equal contribution)

• J. Wang, A.K. Sahu, G. Joshi and S. Kar, Exploring the Error-Runtime Trade-off in Decentralized Optimization, In Proceedings of 54th Annual Asilomar Conference on Signals, Systems and Computers, 2020, Pacific Grove, CA

• T. Li, A.K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar and V. Smith, Federated Optimization in Heterogeneous Networks, In Proceedings of MLSys 2020. (An abridged version appeared in ICML 2019 AMTL Workshop)

• T. Li, A.K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar and V. Smith, Fed-DANE: A Federated Newton-Type Method, In Proceedings of 53rd Annual Asilomar Conference on Signals, Systems and Computers, 2019, Pacific Grove, CA

• R.Xin, A.K. Sahu, U.A. Khan and S. Kar, Distributed empirical risk minimization over directed graphs, In Proceedings of 53rd Annual Asilomar Conference on Signals, Systems and Computers, 2019, Pacific Grove, CA

• R.Xin, A.K. Sahu, U.A. Khan and S. Kar, Distributed stochastic optimization with gradient tracking over strongly-connected networks, In Proceedings of 58th IEEE Conference on Decision and Control, CDC 2019

• A.K. Sahu, M. Zaheer and S. Kar, Towards Gradient Free and Projection Free Stochastic Optimization, In Proceedings of 22nd International Conference on Artificial Intelligence and Statistics(AISTATS)

• A.K. Sahu, D. Jakovetic, D. Bajovic and S. Kar, Communication Efficient Distributed Estimation over Directed Random Graphs, In Proceedings of IEEE EUROCON 2019, 18th International Conference on Smart Technologies

• A.K. Sahu, D.Jakovetic, D. Bajovic and S. Kar, “Non-Asymptotic Rates for Communication Efficient Distributed Zeroth Order Strongly Convex Optimization”, In Proceedings of IEEE Global Conference on Signal and Information Processing(GlobalSIP), 2018

• D.Jakovetic, D. Bajovic, A.K. Sahu and S. Kar, “Convergence rates for distributed stochastic optimization over random networks”, In Proceedings of 57th IEEE Conference on Decision and Control, CDC 2018

• A.K. Sahu, D.Jakovetic, D. Bajovic and S. Kar, “Distributed Zeroth Order Optimization Over Random Networks: A Kiefer-Wolfowitz Stochastic Approximation Approach”, In Proceedings of 57th IEEE Conference on Decision and Control, CDC 2018

• D. Bajovic, D.Jakovetic, A.K. Sahu and S. Kar, “Large Deviations for Products of Non-i.i.d. Stochastic Matrices with Application to Distributed Detection”, In Proceedings of International Symposium on Information Theory, ISIT 2018

• A.K. Sahu, D.Jakovetic and S. Kar, “CREDO: A Communication-Efficient Distributed Estimation Algorithm”, In Proceedings of International Symposium on Information Theory, ISIT 2018

• Z. Jiang, J. Francis, A.K. Sahu, S. Munir, C. Shelton, A. Rowe and M. Berges, Data-driven Thermal Model Inference with ARMAX, in Smart Environments, based on Normalized Mutual Information, In Proceedings of American Control Conference, ACC 2018

• A.K. Sahu and S. Kar, “Dist-Hedge: A Partial Information Setting Based Distributed Non-Stochastic Sequence Prediction Algorithm”, In Proceedings of IEEE Global Conference on Signal and Information Processing(GlobalSIP), 2017

• A.K. Sahu and S. Kar, “Distributed Sequence Prediction: A consensus+innovations approach”, In Proceedings of IEEE Global Conference on Signal and Information Processing(GlobalSIP), 2016

• S. Kar, R. Negi, M. Mahzoon and A.K. Sahu, “Queue-based Broadcast Gossip Algorithm for Consensus”, Accepted. To appear in Proceedings of 54th Annual Allerton Conference on Communication, Control, and Computing, 2016

• A.K. Sahu and S. Kar, “Distributed Recursive Composite Hypothesis Testing: Imperfect Communication”, International Symposium on Information Theory, ISIT 2016

• A.K. Sahu and S. Kar, “Distributed Generalized Likelihood Ratio Tests: Fundamental Limits and Tradeoffs”,International Conference on Acoustics, Speech and Signal Processing(ICASSP), 2016

• A.K. Sahu and S. Kar, “Distributed Sequential Detection for Gaussian Binary Hypothesis Testing : Heterogeneous Networks”, 48th Annual Asilomar Conference on Signals, Systems and Computers, 2014, Pacific Grove, CA

Invited Talks

• “Distributed Inference in Large-scale Stochastic Systems: Imperfect Communication” Department of Electronics and Electrical Communications Engineering, IIT Kharagpur, August 2016

• “Distributed Weighted Non-linear least Squares and Recursive Composite Hypothesis Testing: A consensus+innovations approach”, Mcgill University, July 2017

• “Distributed Recursive Composite Hypothesis Testing: A consensus+innovations approach”, IFORS 2017, July 2017

• “Online Learning in Large Scale Stochastic Systems”, IBM India Research Lab, August 2017