Analytics and Online Optimization at Scale for Cellular Networks
NSF Project CNS (2107037 / 2106933)
Project Summary
Cellular networks have become one of the critical infrastructures for society, with users expecting reliable connectivity and performance. Behind the scenes, operating these networks require updating hundreds of parameters at time scales ranging from hours to weeks, which is extremely costly and inefficient for engineers at the network operations center. Further, when failures or inefficient performance occurs, detecting and isolating the root causes is again a challenging, but critical task. This proposal focuses on efficiently operating these networks and developing tools to detect anomalies, both using machine learning techniques. The goal of this proposal is to develop algorithms based on online learning, Bayesian optimization and deep learning for parameter tuning and anomaly detection. Building on partnerships with major cellular providers and the use of real data-traces and testbeds, our algorithms and approaches will have real-world impact.
Current PIs
Sanjay Shakkottai (UT Austin)
Ness Shroff (Ohio State University)
Publications
“Provably Efficient Model-Free Algorithms for Non-stationary CMDPs,” H. Wei, A. Ghosh, N. B. Shroff, L. Ying, and X. Zhou, AISTATS, Valencia, Spain, Apr. 2023.
“Achieving Sub-linear Regret in Infinite Horizon Average Reward Constrained MDP with Linear Function Approximation,” A. Ghosh, X. Zhou, and N. B. Shroff, ICLR, Kigali, Rwanda, May 2023.
“Learning in Constrained Markov Decision Processes,” R. Singh, A. Gupta, and N. B. Shroff, IEEE Trans on Control of Network Systems, vol. 10, no. 1, pp. 441-453, Mar. 2023.
"Collaborative Multi-Agent Heterogeneous Multi-Armed Bandits", R. Chawla, D. Vial, S. Shakkottai and R. Srikant. Proceedings of the 40th International Conference on Machine Learning (ICML), Honolulu, HI July 2023.
“Near-optimal Adversarial Reinforcement Learning with Switching Costs,” M. Shi, Y. Liang, and N. B. Shroff, ICLR, Kigali, Rwanda, May 2023.
“Robust Multi-Agent Bandits Over Undirected Graphs”, D. Vial, S. Shakkottai and R. Srikant. Proceedings of the ACM Sigmetrics Conference on Measurement and Modeling of Computer Systems (Sigmetrics), Orlando, FL, June 2023.
“Provably Efficient Model-Free Constrained RL with Linear Function Approximation,” A. Ghosh, X. Zhou, and N. B. Shroff, NeurIPS, New Orleans, Louisiana, USA, Nov. 2022.
“Interference Constrained Beam Alignment for Time-Varying Channels via Kernelized Bandits,” Y. Deng, X. Zhou, A. Ghosh, A. Gupta, and N. B. Shroff, IEEE WiOpt’22, Turin, Italy, Sep. 2022.
“Weighted Gaussian Process Bandits for Non-stationary Environments,” Y. Deng, X. Zhou, B. Kim, A. Tewari, A. Gupta, and N. B. Shroff, AISTATS, March 2022.
“Asymptotically-Optimal Gaussian Bandits with Side Observations”, A. Atsidakou, O. Papadigenopoulos, C. Caramanis, S. Sanghavi and S. Shakkottai. Proceedings of the 39th International Conference on Machine Learning (ICML), Baltimore, MD, July 2022.
"Bandit Learning-based Online User Clustering and Selection for Cellular Networks” I. Tariq, K. Patel, T. Novlan, S. Akoum, M. Majmundar, G. de Veciana and S. Shakkottai, IEEE WiOpt’22, Turin, Italy, Sep. 2022.
Broader Impacts Activities
Shroff co-chaired the Future Directions Workshop on Wireless Communications: XG and Beyond
Shakkottai conducted a short course on causal inference at Georgia Tech
Several presentations to industry partners on research outcomes through the 6G@UT program and individual meetings with companies by both PIs
Collaborations with AT&T Labs researchers on creating tools for cellular network management