CORTEXA
← Browse
arxivstat.MLcs.AIcs.LG2026-07-05

Fixed-Confidence Best-Arm Identification for Causal Mediation Analysis

Harsh Shrivastava, Yuta Kawakami, Junpei Komiyama, Jin Tian

This paper studies the problem of identifying the treatment that maximizes the expected natural direct potential outcome (NDPO), which captures the potential outcome of an intervention while excluding the pathway transmitted through a mediator that researchers may wish to remove from evaluation. We first establish population-level identification of the expected NDPO in a causal bandit setting using observable interventional distributions. We then develop a fixed-confidence best-arm identification (BAI) algorithm based on the Track-and-Stop (TaS) framework, employing a cutting-set method to solve the resulting semi-infinite optimization problem. The proposed algorithm achieves sample-efficient identification with a high-probability correctness guarantee. We prove that it satisfies $δ$-correctness and asymptotic optimality. Finally, we validate the approach through empirical evaluations on a large-scale real-world advertising dataset (IPinYou).

View free PDFSource page

Related papers

arxivcs.LGcs.AIstat.ML2026-07-12

Learning from Local Walks on Dynamic Graphs with Bandit Feedback

Sourav Chakraborty, Amit Kiran Rege, Claire Monteleoni, Lijun Chen

We study stochastic multi-armed bandits on dynamic graphs, where arms correspond to the vertices of a network with time-varying edges. In this setting, the learner is restricted to local movement, selecting only its current node or an immediate neighbor at each round. This constr…

View free PDFSource page
arxivcs.CLcs.AIcs.LGstat.ML2026-07-05

CausalGame: Benchmarking Causal Thinking of LLM Agents in Games

Zhenhao Chen, Yongqiang Chen, Chenxi Liu, Junchi Yu, Xiangchen Song, Zijian Li, et al.

Building AI Scientist agents with Large Language Models (LLMs) has recently attracted growing attention. Since scientific discovery fundamentally relies on uncovering causal relationships from observations, the capability of causal thinking, i.e., distinguishing causation from co…

View free PDFSource page
arxivcs.LGcs.AIstat.ML2026-06-29

Pessimism's Paradox: Conservative Offline Training Amplifies Reward Hacking During Online Adaptation in Reasoning Models

Subramanyam Sahoo, Aman Chadha, Vinija Jain, Divya Chaudhary

Conservative offline training is widely advocated as a safe foundation for subsequent online adaptation: if a policy stays close to well-supported behaviour, the argument goes, it is less likely to exploit imperfections in a learned reward model. We challenge this intuition empir…

View free PDFSource page
arxivstat.MLcs.AIcs.LG2026-06-28

Self-Organized Conformal Prediction: Reducing Regional Coverage Gaps with Unsupervised Group Discovery

Louis Berthier, Ahmed Shokry, Maxime Moreaud, Guillaume Ramelet, Aymeric Dieuleveut

Conformal prediction guarantees marginal coverage, but pooled calibration averages over heterogeneous regions and can mask regional undercoverage in safety-critical subgroups. We introduce Self-Organized Conformal Prediction (SOCP), a calibration scheme that discovers input-space…

View free PDFSource page
arxivcs.LGcs.AIstat.ML2026-07-13

Are we Merging the Right Models? Impact of Expert Training Duration on Model Merging for LLMs

Nikita Kozodoi, Zainab Afolabi, Jack Butler

Multi-task model merging combines separately trained expert models into a single model that handles all tasks without co-training. Standard practice merges experts at their optimal validation loss. We challenge this convention by systematically studying how training duration of d…

View free PDFSource page
arxivstat.MEcs.AIcs.LGstat.ML2026-07-15

Verifying formulas for interventional distributions

Francesco Freni, Leonard Henckel, Sebastian Weichwald

We formalize verification in causal graphical models: deciding whether a given observational formula identifies a target interventional distribution. This opens a problem complementary to identification, asking not whether any identifying formula exists, but whether the given for…

View free PDFSource page