AISTATS 2025 论文列表

Individuals often aim to reverse undesired outcomes in interactions with automated systems, like loan denials, by either implementing system-recommended actions (recourse), or manipulating their features. While providing recourse benefits users and enhances system utility, it also provides information about the decision process that can be used for more effective strategic manipulation, especially when the individuals collectively share such information with each other. We show that this tension leads rational utility-maximizing systems to frequently withhold recourse, resulting in decreased population utility, particularly impacting sensitive groups. To mitigate these effects, we explore the role of recourse subsidies, finding them effective in increasing the provision of recourse actions by rational systems, as well as lowering the potential social cost and mitigating unfairness caused by recourse withholding.

We study the problem of representational transfer in offline Reinforcement Learning (RL), where a learner has access to episodic data from a number of source tasks collected a priori, and aims to learn a shared representation to be used in finding a good policy for a target task. Unlike in online RL where the agent interacts with the environment while learning a policy, in the offline setting there cannot be such interactions in either the source tasks or the target task; thus multi-task offline RL can suffer from incomplete coverage. We propose an algorithm to compute pointwise uncertainty measures for the learnt representation in low-rank MDPs, and establish a data-dependent upper bound for the suboptimality of the learnt policy for the target task. Our algorithm leverages the collective exploration done by source tasks to mitigate poor coverage at some points by a few tasks, thus overcoming the limitation of needing uniformly good coverage for a meaningful transfer by existing offline algorithms. We complement our theoretical results with empirical evaluation on a rich-observation MDP which requires many samples for complete coverage. Our findings illustrate the benefits of penalizing and quantifying the uncertainty in the learnt representation.

Causal discovery can be computationally demanding for large numbers of variables. If we only wish to estimate the causal effects on a small subset of target variables, we might not need to learn the causal graph for all variables, but only a small subgraph that includes the targets and their adjustment sets. In this paper, we focus on identifying causal effects between target variables in a computationally and statistically efficient way. This task combines causal discovery and effect estimation, aligning the discovery objective with the effects to be estimated. We show that definite non-ancestors of the targets are unnecessary to learn causal relations between the targets and to identify efficient adjustments sets. We sequentially identify and prune these definite non-ancestors with our Sequential Non-Ancestor Pruning (SNAP) framework, which can be used either as a preprocessing step to standard causal discovery methods, or as a standalone sound and complete causal discovery algorithm. Our results on synthetic and real data show that both approaches substantially reduce the number of independence tests and the computation time without compromising the quality of causal effect estimations.

Stochastic compositional minimax problems are prevalent in machine learning, yet there exist only limited established findings on the convergence of this class of problems. In this paper, we propose a formal definition of the stochastic compositional minimax problem, which involves optimizing a minimax loss with a compositional structure either in primal, dual, or both primal and dual variables. We introduce a simple yet effective algorithm, stochastically Corrected stOchastic gradient Descent Ascent (CODA), which is a primal-dual type algorithm with compositional correction steps, and establish its convergence rate in the aforementioned three settings. We also propose a variance reduced variant, CODA+, which achieves the best-known rate on nonconvex-strongly-concave and nonconvex-concave compositional minimax problems. This work initiates the theoretical study of the stochastic compositional minimax problem in various settings and may inform modern machine learning scenarios such as domain adaptation or robust model-agnostic meta-learning.

Acyclic directed mixed graphs (ADMG) – graphs that contain both directed and bidi- rected edges but no directed cycles – are used to model causal and conditional independence relationships between a set of random vari- ables in the presence of latent or unmeasured variables. Bow-free ADMGs, Arid ADMGs, and Ancestral ADMGs (AADMG) are three widely studied classes of ADMGs where each class is contained in the previously mentioned class. There are a number of published meth- ods – primarily heuristic ones – to find score- maximizing AADMGs from data. Bow-free and Arid ADMGs can model certain equal- ity restrictions – such as Verma constraints – between observed variables that maximal AADMGs cannot. In this work, we develop the first exact methods – based on integer programming – to find score-maximizing Bow- free and Arid ADMGs. Our methods work for data that follows a continuous Gaussian distri- bution and for scores that linearly decompose into the sum of scores of c-components of an ADMG. To improve scaling, we develop an effective linear-programming based heuris- tic that yields solutions with high parent set sizes and/or large districts. We show that our proposed algorithms obtain better scores than other state-of-the-art methods and re- turn graphs that have excellent fits to data.