Strong Representation for ECE Faculty at ICML

The University of Maryland had a strong showing at the 40^th International Conference on Machine Learning (ICML) 2023, held last month in Honolulu, Hawaii. The Outstanding Paper Award was presented to UMD researchers John Kirchenbauer, Jonas Geiping, Yuxin Wen, Jonathan Katz, Ian Miers, and Tom Goldstein (Katz and Goldstein are affiliated with ECE). Their paper, titled “A Watermark for Large Language Models” focuses on embedding watermarks into large language models to help identify the source of specific outputs generated by the model.

Other papers presented by ECE faculty include:

Revisiting the Linear-Programming Framework for Offline RL with General Function Approximation

Asuman Ozdaglar, Sarath Pattathil, Jiawei Zhang, Kaiqing Zhang (ECE)

STEERING : Stein Information Directed Exploration for Model-Based Reinforcement Learning

Souradip Chakraborty (UMD), Amrit Bedi (UMD), Alec Koppel, Mengdi Wang, Furong Huang (UMD), Dinesh Manocha (ECE)

Robust Counterfactual Explanations for Neural Networks With Probabilistic Guarantees

Faisal Hamman, Erfaun Noorani, Saumitra Mishra, Daniele Magazzeni, Sanghamitra Dutta (ECE)

Beyond Exponentially Fast Mixing in Average-Reward Reinforcement Learning via Multi-Level Monte Carlo Actor-Critic

Wesley A. Suttle, Amrit Bedi (UMD), Bhrij Patel, Brian Sadler, Alec Koppel, Dinesh Manocha (ECE)

Partially Observable Multi-agent RL with (Quasi-)Efficiency: The Blessing of Information Sharing

Xiangyu Liu, Kaiqing Zhang (ECE)

Fed-CBS: A Heterogeneity-Aware Client Sampling Mechanism for Federated Learning via Class-Imbalance Reduction