D'ya like dags? a survey on structure learning and causal discovery

MJ Vowels, NC Camgoz, R Bowden - ACM Computing Surveys, 2022 - dl.acm.org
Causal reasoning is a crucial part of science and human intelligence. In order to discover
causal relationships from data, we need structure discovery methods. We provide a review …

Towards robust and adaptive motion forecasting: A causal representation perspective

Y Liu, R Cadei, J Schweizer… - Proceedings of the …, 2022 - openaccess.thecvf.com
Learning behavioral patterns from observational data has been a de-facto approach to
motion forecasting. Yet, the current paradigm suffers from two shortcomings: brittle under …

Local permutation tests for conditional independence

I Kim, M Neykov, S Balakrishnan… - The Annals of …, 2022 - projecteuclid.org
Local permutation tests for conditional independence Page 1 The Annals of Statistics 2022, Vol.
50, No. 6, 3388–3414 https://doi.org/10.1214/22-AOS2233 © Institute of Mathematical Statistics …

Engression: extrapolation through the lens of distributional regression

X Shen, N Meinshausen - … of the Royal Statistical Society Series …, 2024 - academic.oup.com
Distributional regression aims to estimate the full conditional distribution of a target variable,
given covariates. Popular methods include linear and tree ensemble based quantile …

Conditional independence testing under misspecified inductive biases

F Maia Polo, Y Sun, M Banerjee - Advances in Neural …, 2023 - proceedings.neurips.cc
Conditional independence (CI) testing is a fundamental and challenging task in modern
statistics and machine learning. Many modern methods for CI testing rely on powerful …

Does the Markov decision process fit the data: Testing for the Markov property in sequential decision making

C Shi, R Wan, R Song, W Lu… - … Conference on Machine …, 2020 - proceedings.mlr.press
The Markov assumption (MA) is fundamental to the empirical validity of reinforcement
learning. In this paper, we propose a novel Forward-Backward Learning procedure to test …

Algorithm-agnostic significance testing in supervised learning with multimodal data

L Kook, AR Lundborg - Briefings in Bioinformatics, 2024 - academic.oup.com
Motivation Valid statistical inference is crucial for decision-making but difficult to obtain in
supervised learning with multimodal data, eg combinations of clinical features, genomic …

Statistical testing under distributional shifts

N Thams, S Saengkyongam, N Pfister… - Journal of the Royal …, 2023 - academic.oup.com
We introduce statistical testing under distributional shifts. We are interested in the hypothesis
P*∈ H 0 for a target distribution P*, but observe data from a different distribution Q*. We …

A robust test for the stationarity assumption in sequential decision making

J Wang, C Shi, Z Wu - International Conference on Machine …, 2023 - proceedings.mlr.press
Reinforcement learning (RL) is a powerful technique that allows an autonomous agent to
learn an optimal policy to maximize the expected return. The optimality of various RL …

K-nearest-neighbor local sampling based conditional independence testing

S Li, Y Zhang, H Zhu, C Wang, H Shu… - Advances in …, 2024 - proceedings.neurips.cc
Conditional independence (CI) testing is a fundamental task in statistics and machine
learning, but its effectiveness is hindered by the challenges posed by high-dimensional …