8 papers
CAREAgent: Clinical Agent with Structured Reasoning and Tool-Integrated for Order Generation
Ruihui Hou, Ziyue Huai, Chennuo Zhang +5
Clinical order generation serves as a critical bridge between clinical decision-making and real-world practice, translating medical decisions into concrete and executable orders. E…
The Representation-Rationalizability Tradeoff in Reward Learning
Jing Dong, Yaoliang Yu, Pascal Pourpart
In RLHF, each training example contains a prompt and two candidate responses , and annotators provide pairwise preferences between these responses. The learning problem i…
A Unified Framework for Gradient Aggregation in Multi-Objective Optimization
Zeou Hu, Kelvin Ho, Yaoliang Yu
Many machine learning problems involve multiple inherent trade-offs that are best addressed by gradient-based multi-objective optimization (MOO) algorithms. Existing methods are of…
Efficient Bilevel Optimization with KFAC-Based Hypergradients
Disen Liao, Felix Dangel, Yaoliang Yu
Bilevel optimization (BO) is widely applicable to many machine learning problems. Scaling BO, however, requires repeatedly computing hypergradients, which involves solving inverse…
Last-iterate Convergence in Regularized Graphon Mean Field Game
Jing Dong, Baoxiang Wang, Yaoliang Yu
To model complex real-world systems, such as traders in stock markets, or the dissemination of contagious diseases, graphon mean-field games (GMFG) have been proposed to model many…
A Comprehensive Framework for Analyzing the Convergence of Adam: Bridging the Gap with SGD
Ruinan Jin, Xiao Li, Yaoliang Yu +1
Adaptive Moment Estimation (Adam) is a cornerstone optimization algorithm in deep learning, widely recognized for its flexibility with adaptive learning rates and efficiency in han…
Uncoupled and Convergent Learning in Monotone Games under Bandit Feedback
Jing Dong, Baoxiang Wang, Yaoliang Yu
We study the problem of no-regret learning algorithms for general monotone and smooth games and their last-iterate convergence properties. Specifically, we investigate the problem…
Convergence to Nash Equilibrium and No-regret Guarantee in (Markov) Potential Games
Jing Dong, Baoxiang Wang, Yaoliang Yu
In this work, we study potential games and Markov potential games under stochastic cost and bandit feedback. We propose a variant of the Frank-Wolfe algorithm with sufficient explo…