Zobrazeno 1 - 10
of 126
pro vyhledávání: '"Shalit, Uri"'
Merging datasets across institutions is a lengthy and costly procedure, especially when it involves private information. Data hosts may therefore want to prospectively gauge which datasets are most beneficial to merge with, without revealing sensitiv
Externí odkaz:
http://arxiv.org/abs/2409.07215
In many reinforcement learning (RL) applications one cannot easily let the agent act in the world; this is true for autonomous vehicles, healthcare applications, and even some recommender systems, to name a few examples. Offline RL provides a way to
Externí odkaz:
http://arxiv.org/abs/2407.00806
Vital signs are crucial in intensive care units (ICUs). They are used to track the patient's state and to identify clinically significant changes. Predicting vital sign trajectories is valuable for early detection of adverse events. However, conventi
Externí odkaz:
http://arxiv.org/abs/2403.18668
Publikováno v:
PMLR 202 (2023) 26599-26618
Estimating heterogeneous treatment effects from observational data is a crucial task across many fields, helping policy and decision-makers take better actions. There has been recent progress on robust and efficient methods for estimating the conditi
Externí odkaz:
http://arxiv.org/abs/2304.10577
Learned classifiers should often possess certain invariance properties meant to encourage fairness, robustness, or out-of-distribution generalization. However, multiple recent works empirically demonstrate that common invariance-inducing regularizers
Externí odkaz:
http://arxiv.org/abs/2211.15724
Autor:
Tennenholtz, Guy, Merlis, Nadav, Shani, Lior, Mannor, Shie, Shalit, Uri, Chechik, Gal, Hallak, Assaf, Dalal, Gal
We present the problem of reinforcement learning with exogenous termination. We define the Termination Markov Decision Process (TerMDP), an extension of the MDP framework, in which episodes may be interrupted by an external non-Markovian observer. Th
Externí odkaz:
http://arxiv.org/abs/2205.15376
Autor:
Jesson, Andrew, Douglas, Alyson, Manshausen, Peter, Solal, Maëlys, Meinshausen, Nicolai, Stier, Philip, Gal, Yarin, Shalit, Uri
Estimating the effects of continuous-valued interventions from observational data is a critically important task for climate science, healthcare, and economics. Recent work focuses on designing neural network architectures and regularization function
Externí odkaz:
http://arxiv.org/abs/2204.10022
Autor:
Jesson, Andrew, Tigas, Panagiotis, van Amersfoort, Joost, Kirsch, Andreas, Shalit, Uri, Gal, Yarin
Estimating personalized treatment effects from high-dimensional observational data is essential in situations where experimental designs are infeasible, unethical, or expensive. Existing approaches rely on fitting deep models on outcomes observed for
Externí odkaz:
http://arxiv.org/abs/2111.02275
We consider the problem of using expert data with unobserved confounders for imitation and reinforcement learning. We begin by defining the problem of learning from confounded expert data in a contextual MDP setup. We analyze the limitations of learn
Externí odkaz:
http://arxiv.org/abs/2110.06539
Publikováno v:
PMLR 139 (2021) 4829-4838
We study the problem of learning conditional average treatment effects (CATE) from high-dimensional, observational data with unobserved confounders. Unobserved confounders introduce ignorance -- a level of unidentifiability -- about an individual's r
Externí odkaz:
http://arxiv.org/abs/2103.04850