Publications
Preprints:
Detection of sensor attacks in cyber-physical systems: algorithm and guarantees; Souvik Das, Avishek Ghosh and Debasish Chatterjee, 2024
Learning from Mixtures with Applications to Heterogeneous Federated Learning; Harsh Vardhan, Avishek Ghosh and Arya Mazumdar, 2023.
Journal Papers:
An Improved Algorithm for Clustered Federated Learning; Harshvardhan, Avishek Ghosh and Arya Mazumdar --- Transactions in Machine Learning Research (TMLR), 2024.
Decentralized Competing Bandits In Non-Stationary Matching Markets; Avishek Ghosh, Abishek Sankararaman, Kannan Ramchandran, Tara Javidi and Arya Mazumdar-- IEEE Transactions on Information Theory, 2023.
Model Selection for Generic Contextual Bandits; Avishek Ghosh, Abishek Sankararaman, Kannan Ramchandran -- IEEE Transactions on Information Theory, 2023.
Understanding and Control of Zener Pinning via Phase Field and Ensemble Learning; Sukriti Manna, Henry Chan, Avishek Ghosh, Tamoghna Chakrabarti and Subramanian Sankaranarayanan -- Computational Materials Science, 2023
An Efficient Framework for Clustered Federated Learning (long version) -- Avishek Ghosh, Jichan Chung, Dong Yin, Kannan Ramchandran -- IEEE Transactions on Information Theory, 2022.
Max-Affine Regression: Parameter Estimation for Gaussian Designs; Avishek Ghosh, Ashwin Pananjady, Aditya Guntuboyina, Kannan Ramchandran, IEEE Transactions on Information Theory, 2022.
Communication-Efficient and Byzantine-Robust Distributed Learning with Error Feedback; Avishek Ghosh, Raj Kumar Maity, Swanand Kadhe, Arya Mazumdar and Kannan Ramchandran, IEEE Journal on Selected Areas in Information Theory, 2021.
Asynchronous Stochastic Approximation Based Learning Algorithms for As-You-Go Deployment of Wireless Relay Networks along a Line; Arpan Chattopadhyay, Avishek Ghosh , Anurag Kumar. IEEE Transactions on Mobile Computing, 2018
Measurement Based As-You-Go Deployment of Two-Connected Wireless Relay Networks; Avishek Ghosh, Arpan Chattopadhyay, Arora Anish, Anurag Kumar. ACM Transactions on Sensor Networks, 2017.
Integral Value Transformations: A Class of Affine Discrete Dynamical Systems and an Application; Sk. S. Hassan, P. Pal Choudhury, B. K. Nayak, Avishek Ghosh, J. Banerjee, Journal of Advanced Research in Applied Mathematics (JARAM), 2011.
A novel genetic algorithm to solve travelling salesman problem and blocking flow shop scheduling problem; A. Chowdhury, A. Ghosh, S. Sinha, S. Das and A Ghosh International Journal Bio-Inspired Computation (IJBIC) (2013).
Conference Papers:
Agnostic Learning of Mixed Linear Regressions with EM and AM Algorithms; Avishek Ghosh and Arya Mazumdar; International Conference on Machine Learning (ICML), 2024, Vienna, Austria
PairNet: Training with Observed Pairs to Estimate Individual Treatment Effect; Lokesh N, Pranava Singhal, Avishek Ghosh and Sunita Sarawagi; International Conference on Machine Learning (ICML), 2024, Vienna, Austria
Explore-then-Commit Algorithms for Decentralized Two-sided Matching Markets; Tejas Pagare, Avishek Ghosh-- IEEE International Symposium on Information Theory (ISIT) 2024, Athens, Greece
DIST-CURE: A Robust Distributed Learning Algorithm with Cubic Regularized Newton; Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar-- IEEE International Symposium on Information Theory (ISIT) 2024, Athens, Greece
Detection of False Data Injection Attacks in Cyber-Physical Systems; Souvik Das, Avishek Ghosh, Debasish Chatterjee-- IEEE International Symposium on Information Theory (ISIT) 2024, Athens, Greece
Optimal Compression of Unit Norm Vectors in the High Distortion Regime; Heng Zhu, Avishek Ghosh, Arya Mazumdar, -- IEEE International Symposium on Information Theory (ISIT) 2023, Taipei, Taiwan
Exploration in Linear Bandits with Rich Action Sets and its Implications for Inference; Debangshu Banerjee, Avishek Ghosh, Sayak Raychowdhury and Aditya Gopalan -- International Conference on Artificial Intelligence and Statistics (AISTATS) 2023, Valencia, Spain
Breaking the $\sqrt{T}$ Barrier: Instance-Independent Logarithmic Regret in Stochastic Contextual Linear Bandits; Avishek Ghosh and Abishek Sankararaman, International Conference on Machine Learning (ICML), 2022, Baltimore, USA
On Learning Mixture of Linear Regressions in the Non-Realizable Setting; Avishek Ghosh, Arya Mazumdar, Soumyabrata Pal and Rajat Sen, International Conference on Machine Learning (ICML), 2022, Baltimore, USA
Model Selection in Reinforcement Learning with General Function Approximations; Avishek Ghosh and Sayak Ray Chowdhury-- European Conference on Machine Learning (ECML-PKDD), 2022
Multi-Agent Heterogeneous Stochastic Linear Bandits; Avishek Ghosh, Abishek Sankararaman, Kannan Ramchandran-- European Conference on Machine Learning (ECML-PKDD), 2022
LocalNewton: Reducing Communication Rounds for Distributed Learning --- Vipul Gupta, Avishek Ghosh, Michal Derezinski, Rajiv Khanna, Kannan Ramchandran, Michael W Mahoney-- Uncertainty in Artificial Intelligence (UAI), 2021
Problem-Complexity Adaptive Model Selection for Stochastic Linear Bandits -- Avishek Ghosh, Abishek Sankararaman, Kannan Ramchandran-- International Conference on Artificial Intelligence and Statistics (AISTATS), 2021 (https://arxiv.org/pdf/2006.02612.pdf)
An Efficient Framework for Clustered Federated Learning -- Avishek Ghosh, Jichan Chung, Dong Yin, Kannan Ramchandran -- Annual Conference on Neural Information Processing Systems (NeurIPS), 2020 (https://arxiv.org/abs/2006.04088)
Distributed Newton Can Communicate Less and Resist Byzantine Workers -- Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar -- Annual Conference on Neural Information Processing Systems (NeurIPS), 2020. (https://arxiv.org/abs/2006.0873)
Max-affine Regression with Universal Parameter Estimation for Small-ball Designs; Avishek Ghosh, Ashwin Pananjady, Aditya Guntuboyina, Kannan Ramchandran -- IEEE International Symposium on Information Theory (ISIT), 2020, LA, USA.
Communication Efficient Distributed Approximate Newton Method; Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar, Kannan Ramchandran -- IEEE International Symposium on Information Theory (ISIT), 2020, LA, USA.
Communication Efficient and Byzantine Tolerant Distributed Learning; Avishek Ghosh, Raj Kumar Maity, Swanand Kadhe, Arya Mazumdar, Kannan Ramchandran -- IEEE International Symposium on Information Theory (ISIT), 2020, LA, USA.
Some Performance Guarantees of Global LASSO with Local Assumptions for Convolutional Sparse Design Matrices; Avishek Ghosh, Kannan Ramchandran -- IEEE International Symposium on Information Theory (ISIT), 2020, LA, USA.
Alternating Minimization Converges Super-linearly for Mixed Linear Regression; Avishek Ghosh and Kannan Ramchandran -- International Conference on Artificial Intelligence and Statistics (AISTATS), 2020, Sicily, Italy.
Matching Observations to Distributions: Efficient Estimation via Sparsified Hungarian Algorithm; Sinho Chewi, Forest Yang, Avishek Ghosh, Abhay Parekh and Kannan Ramchandran -- Allerton Conference on Communication, Control, and Computing, 2019, IL, USA.
Online Scoring with Delayed Information: A Convex Optimization Viewpoint ; Avishek Ghosh, Kannan Ramchandran -- Allerton Conference on Communication, Control, and Computing (2018) , IL, USA.
Faster Data-access in Large-scale Systems: Network-scale Latency Analysis under General Service-time Distributions ; Avishek Ghosh, Kannan Ramchandran -- Allerton Conference on Communication, Control, and Computing (2018) , IL, USA.
Misspecified Linear Bandits ; Avishek Ghosh, Sayak Ray Chowdhury and Aditya Gopalan -- AAAI Conference on Artificial Intelligence (AAAI), 2017, Sanfrancisco, USA.
As-you-go deployment of a 2-connected wireless relay network for sensor-sink interconnection; Avishek Ghosh, Arpan Chattopadhyay, Anish Arora, and Anurag Kumar; IEEE Signal Processing and Communications (IEEE-SPCOM), 2014, Bangalore, India.
Impromptu Deployment of Wireless Relay Networks: Experiences Along a Forest Trail; Arpan Chattopadhyay, Avishek Ghosh, Akhila S Rao, Bharat Dwivedi, SVR Anand, Marceau Coupechoux, Anurag Kumar -- IEEE Mobile Ad Hoc and Sensor Systems (IEEE-MASS), 2014.
Linear phase low pass FIR filter design using Genetic Particle Swarm Optimization with dynamically varying neighbourhood technique ; Avishek Ghosh, Arnab Ghosh, Arkabandhu Chowdhury, Amit Konar, Eunjin Kim, Atulya K Nagar -- IEEE Congress on Evolutionary Computation (IEEE-CEC), 2012.
Multi-robot cooperative box-pushing problem using multi-objective particle swarm optimization Technique ; Arnab Ghosh, Avishek Ghosh, A Konar, R Janarthanan -- World Congress on Information and Communication Technologies (WICT), 2012.
An evolutionary approach to drug-design using quantam binary particle swarm optimization algorithm; Avishek Ghosh, Arnab Ghosh, Arkabandhu Chowdhury, Jubin Hazra -- IEEE Students' Conference on Electrical, Electronics and Computer Science, 2012.
Application of Integral Value Transformation (IVT) in a Specialized Computer Network Design; Souvik Naskar, Avishek Ghosh, Pabitra Pal Choudhury -- 99th Indian Science Congress, 2012.
Fractal String Generation and Its Application in Music Composition; Avishek Ghosh, Joydeep Banerjee, Sk. S. Hassan, P. Pal Choudhury. Published in the Proceedings of Mathematical and Statistical Modeling in Innovative Areas (MASTMIA, 2011).
An Evolutionary Approach to Drug-Design Using a Novel Neighbourhood Based Genetic Algorithm ; Arnab Ghosh, Avishek Ghosh, Arkabandhu Chowdhury, Amit Konar -- Priprint: arXiv:1205.6412.
Workshop Papers/Posters:
Two-Sided Bandit Learning in Fully-Decentralized Matching Markets; Tejas Pagare and Avishek Ghosh -- International Conference on Machine Learning (ICML) Workshop on Many Facets of Preference-Based Learning, Hawaii, 2023
Decentralized Competing Bandits In Non-Stationary Matching Markets; Avishek Ghosh, Abishek Sankararaman, Kannan Ramchandran, Tara Javidi and Arya Mazumdar -- International Conference on Machine Learning (ICML) Workshop on Many Facets of Preference-Based Learning, Hawaii, 2023
An improved algorithm for Clustered Federated Learning; Harshvardhan, Avishek Ghosh and Arya Mazumdar -- International Conference on Machine Learning (ICML) Workshop on Federated Learning and Analytics, Hawaii, 2023
Escaping Saddle Points in Distributed Newton's Method with Communication efficiency and Byzantine Resilience; Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar-- International Conference on Machine Learning (ICML) Workshop on Federated Learning and Analytics, Hawaii, 2023
An improved algorithm for Clustered Federated Learning; Harshvardhan, Avishek Ghosh and Arya Mazumdar -- Information Theory and Applications (ITA), Student Poster, 2023, San Diego, USA.
Distributed Newton Can Communicate Less and Resist Byzantine Workers -- Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar-- NSF-TRIPODS Workshop on Communication Efficient Distributed Optimization, 2022
Escaping Saddle Points in Distributed Newton's Method with Communication efficiency and Byzantine Resilience; Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar, Kannan Ramchandran -- NSF-TRIPODS Workshop on Communication Efficient Distributed Optimization, 2022
Model Selection for Finite and Continuous-Armed Stochastic Contextual Bandits-- Avishek Ghosh, Abishek Sankararaman, Kannan Ramchandran -- International Conference on Machine Learning (ICML) Workshop on Theoretical Foundations of Reinforcement Learning and Privacy, 2020, Austria.
An Efficient Framework for Clustered Federated Learning -- Avishek Ghosh, Jichan Chung, Dong Yin, Kannan Ramchandran -- International Conference on Machine Learning (ICML) Workshop on Federated Learning and Privacy, 2020, Austria.
Distributed Newton Can Communicate Less and Resist Byzantine Workers (short version) -- Avishek Ghosh, Raj Kumar Maity, Arya Mazumdar -- International Conference on Machine Learning (ICML) Workshop on Beyond First Order Opt. in ML, 2020, Austria.
Communication-Efficient Byzantine-Robust Distributed Learning-- Avishek Ghosh, Raj Kumar Maity, Swanand Kadhe, Arya Mazumdar and Kannan Ramchandran-- Information Theory and Applications (ITA), Student Poster, 2020, San Diego, USA.
Alternating Minimization for Max-Affine Regression-- Avishek Ghosh, Ashwin Pananjady, Aditya Guntuboyina, Kannan Ramchandran -- Signal Processing with Adaptive Sparse Structured Representations (SPARS), 2019 , Toulouse, France.
Max affine regression: algorithm and analysis-- Avishek Ghosh, Ashwin Pananjady, Aditya Guntuboyina, Kannan Ramchandran, IND-STATS (Innovations in Data and Statistical Sciences), IIT Bombay, 2019, India
Robust Heterogeneous Federated Learning-- Avishek Ghosh, Justin Hong, Dong Yin, Kannan Ramchandran -- International Conference on Machine Learning (ICML) Workshop on Privacy and Security in ML, 2019, LA, USA.
PhD Thesis:
Provable and Efficient Algorithms for Federated, Batch and Reinforcement Learning (see here) --- Avishek Ghosh--- Dept. of EECS, UC Berkeley, May 2021
Masters Thesis:
As-You-Go Deployment of 2 connected Wireless Sensor Networks; Decision Theoretic based design --- Avishek Ghosh--- Dept. of ECE, Indian Institute of Science (IISc) Bangalore, May 2014