On Transportability for Structural Causal Bandits

• URL: http://arxiv.org/abs/2511.17953v1
• Date: Sat, 22 Nov 2025 07:38:11 GMT
• Title: On Transportability for Structural Causal Bandits
• Abstract summary: The structural causal bandit framework provides a graphical characterization for identifying actions that are unable to maximize rewards.<n>We show how priors from the source environments are fused to enhance learning in the deployment setting.<n>The resulting bandit algorithm achieves a sub-linear regret bound with an explicit dependence on informativeness of prior data.
• Score: 12.969694382964052
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Intelligent agents equipped with causal knowledge can optimize their action spaces to avoid unnecessary exploration. The structural causal bandit framework provides a graphical characterization for identifying actions that are unable to maximize rewards by leveraging prior knowledge of the underlying causal structure. While such knowledge enables an agent to estimate the expected rewards of certain actions based on others in online interactions, there has been little guidance on how to transfer information inferred from arbitrary combinations of datasets collected under different conditions -- observational or experimental -- and from heterogeneous environments. In this paper, we investigate the structural causal bandit with transportability, where priors from the source environments are fused to enhance learning in the deployment setting. We demonstrate that it is possible to exploit invariances across environments to consistently improve learning. The resulting bandit algorithm achieves a sub-linear regret bound with an explicit dependence on informativeness of prior data, and it may outperform standard bandit approaches that rely solely on online learning.

Related papers

• Knowledge-Informed Local Causal Discovery of Optimal Adjustment Sets [40.628201358330976]
  Local causal discovery is a scalable alternative to global structure learning.<n>We propose b-LOAD, a knowledge-informed extension of the LOAD algorithm for local discovery of optimal adjustment sets.<n>Results on real-world biological networks show that locally targeted prior knowledge provides the largest gains.
  arXiv  Detail & Related papers  (2026-07-05T18:18:42Z)
• Learning Peer Influence Probabilities with Linear Contextual Bandits [5.887969742827489]
  We study learning peer influence probabilities under a contextual linear bandit framework.<n>We show that a fundamental trade-off can arise between regret minimization and estimation error, characterize all achievable rate pairs, and propose an uncertainty-guided exploration algorithm.
  arXiv  Detail & Related papers  (2025-10-21T22:38:52Z)
• Actionable Counterfactual Explanations Using Bayesian Networks and Path Planning with Applications to Environmental Quality Improvement [4.041834517339835]
  We develop a method for actionable counterfactual explanations.<n>Data is only used to learn a density estimator.<n>We test our algorithm with a real-world Environmental Protection Agency dataset.
  arXiv  Detail & Related papers  (2025-08-04T17:20:50Z)
• Exploiting Edge Features for Transferable Adversarial Attacks in Distributed Machine Learning [54.26807397329468]
  This work explores a previously overlooked vulnerability in distributed deep learning systems.<n>An adversary who intercepts the intermediate features transmitted between them can still pose a serious threat.<n>We propose an exploitation strategy specifically designed for distributed settings.
  arXiv  Detail & Related papers  (2025-07-09T20:09:00Z)
• Interventional Causal Structure Discovery over Graphical Models with Convergence and Optimality Guarantees [0.0]
  We develop a bilevel optimization (Bloom) framework for causal structure learning. Bloom not only provides theoretical support for causal structure discovery from both interventional and observational data, but also aspires to an efficient causal discovery algorithm. It is seen through experiments on both synthetic and real-world datasets that Bloom markedly surpasses other leading learning algorithms.
  arXiv  Detail & Related papers  (2024-08-09T02:22:50Z)
• Reinforcement Learning with Lookahead Information [7.696213902533885]
  We study reinforcement learning problems in which agents observe the reward or transition realizations at their current state before deciding which action to take. Previous work shows that this lookahead information can drastically increase the collected reward. We design provably-efficient learning algorithms able to incorporate lookahead information.
  arXiv  Detail & Related papers  (2024-06-04T12:29:51Z)
• Can Active Sampling Reduce Causal Confusion in Offline Reinforcement Learning? [58.942118128503104]
  Causal confusion is a phenomenon where an agent learns a policy that reflects imperfect spurious correlations in the data. This phenomenon is particularly pronounced in domains such as robotics. In this paper, we study causal confusion in offline reinforcement learning.
  arXiv  Detail & Related papers  (2023-12-28T17:54:56Z)
• Efficient Transfer Learning via Causal Bounds [8.981637739384674]
  We analyze how causal side-information accelerates online learning, and experiments on data reduction.<n>Our analysis precisely characterizes when how causal side-information accelerates online learning, and experiments on data reduction.
  arXiv  Detail & Related papers  (2023-08-07T13:24:50Z)
• Enhancing Multiple Reliability Measures via Nuisance-extended Information Bottleneck [77.37409441129995]
  In practical scenarios where training data is limited, many predictive signals in the data can be rather from some biases in data acquisition. We consider an adversarial threat model under a mutual information constraint to cover a wider class of perturbations in training. We propose an autoencoder-based training to implement the objective, as well as practical encoder designs to facilitate the proposed hybrid discriminative-generative training.
  arXiv  Detail & Related papers  (2023-03-24T16:03:21Z)
• Principled Knowledge Extrapolation with GANs [92.62635018136476]
  We study counterfactual synthesis from a new perspective of knowledge extrapolation. We show that an adversarial game with a closed-form discriminator can be used to address the knowledge extrapolation problem. Our method enjoys both elegant theoretical guarantees and superior performance in many scenarios.
  arXiv  Detail & Related papers  (2022-05-21T08:39:42Z)
• On Covariate Shift of Latent Confounders in Imitation and Reinforcement Learning [69.48387059607387]
  We consider the problem of using expert data with unobserved confounders for imitation and reinforcement learning. We analyze the limitations of learning from confounded expert data with and without external reward. We validate our claims empirically on challenging assistive healthcare and recommender system simulation tasks.
  arXiv  Detail & Related papers  (2021-10-13T07:31:31Z)
• Approximate Regions of Attraction in Learning with Decision-Dependent Distributions [11.304363655760513]
  We analyze repeated risk minimization as the trajectories of the gradient flows of performative risk minimization. We provide conditions to characterize the region of attraction for the various equilibria in this setting. We introduce the notion of performative alignment, which provides a geometric condition on the convergence of repeated risk minimization to performative risk minimizers.
  arXiv  Detail & Related papers  (2021-06-30T18:38:08Z)
• Attribute-Guided Adversarial Training for Robustness to Natural Perturbations [64.35805267250682]
  We propose an adversarial training approach which learns to generate new samples so as to maximize exposure of the classifier to the attributes-space. Our approach enables deep neural networks to be robust against a wide range of naturally occurring perturbations.
  arXiv  Detail & Related papers  (2020-12-03T10:17:30Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.