Learning a response law for continuous decisions from observational time series is central to many real-world decision systems, yet temporal non-stationarity poses a fundamental challenge: as outcomes drift over time, naively pooled models can conflate action effects with shifting baselines and even reverse action rankings. Even when the one-step response is identifiable, unrestricted cross-time generalization remains fundamentally ill-posed. We therefore seek the part of the response that remains stable for decisions even when outcome levels do not. We propose deployable causal action geometry: the action-response shape that remains after anchoring away action-independent level shifts on a selected link scale and profiling out positive amplitude variation. Building on this target, we develop a two-stage orthogonal estimator that first obtains cross-fitted pilot estimates of anchored responses and then performs honest profiled selection over link, representation, and shape classes. The resulting estimator admits oracle guarantees for the restricted target and, under approximate persistence, produces decision rules that remain near optimal in future periods. Experiments on synthetic data and Dominick's support the proposed target and its transfer behavior. In a randomized Taobao A/B test, the deployed system improves GMV and purchase counts by \(1.16\%\) and \(1.70\%\).
Tackling Non-Stationarity in Reinforcement Learning via Causal-Origin Representation
School of Computer Science, Peking University(北京大学计算机学院);Center for Statistical Science, Peking University(北京大学统计科学中心);School of Data Science, Fudan University(复旦大学数据科学学院);Beijing Academy of Artificial Intelligence(北京人工智能研究院)
Beyond Stationarity in Time Series: Discovering Causal Structures and Latent Regimes via Markov Blankets
Univ. Grenoble Alpes(格勒诺布尔大学);CNRS(法国国家科学研究中心);Grenoble INP(格勒诺布尔国立高等工程学院);LIG(信息与信号处理实验室);Sorbonne Université(索邦大学);INSERM(法国国家健康与医学研究院);Institut Pierre Louis d’Epidémiologie et de Santé Publique(皮埃尔·路易流行病学与公共卫生研究所)
Causal-Audit: A Framework for Risk Assessment of Assumption Violations in Time-Series Causal Discovery
Institute for Systems and Robotics, Instituto Superior Técnico(系统与机器人研究所,里斯本高等理工学院);Digital Scholarship at Oxford, University of Oxford(牛津大学数字学术中心);Jet Propulsion Laboratory, California Institute of Technology(加州理工学院喷气推进实验室)
Factored Adaptation for Non-Stationary Reinforcement Learning
City University of Hong Kong(香港城市大学);Carnegie Mellon University(卡内基梅隆大学);Mohamed bin Zayed University of Artificial Intelligence(穆罕默德·本·扎耶德人工智能大学);University of Amsterdam(阿姆斯特丹大学);MIT-IBM Watson AI Lab(麻省理工学院-IBM沃森人工智能实验室)
Stable Time Series Prediction of Enterprise Carbon Emissions Based on Causal Inference
School of Economics and Management, China University of Geosciences (Beijing)(经济管理学院,中国地质大学(北京));MOE Laboratory of Philosophy and Social Sciences for Mineral Resources Security Governance, China University of Geosciences (Beijing)(矿产资源安全保障治理哲学社会科学教育部重点实验室,中国地质大学(北京))
Causal Discovery from Heteroscedastic Stochastic Dynamical Systems under Imperfect Physical Models
Department of Mechanical & Industrial Engineering(机械与工业工程系);Northeastern University(东北大学);Department of Industrial Engineering and Management Sciences(工业工程与管理科学系);Department of Mechanical Engineering(机械工程系);Northwestern University(西北大学)
TTCD:Transformer Integrated Temporal Causal Discovery from Non-Stationary Time Series Data
Department of Information Systems(信息系统系);University of Maryland, Baltimore County(马里兰大学巴尔的摩县分校);Department of Data Science(数据科学系);University of North Texas(北德克萨斯大学);Climate Science Section(气候科学组);Lawrence Livermore National Laboratory(劳伦斯利弗莫尔国家实验室)
M-CaStLe: Uncovering Local Causal Structures in Multivariate Space-Time Gridded Data
Scientific Machine Learning, Center for Computing Research(科学机器学习,计算研究中心);Sandia National Laboratories(沙伊达国家实验室);Zicklin School of Business Baruch College, CUNY(巴鲁克学院Zicklin商学院,纽约大学);Department of Computer Science University of New Mexico(新墨西哥大学计算机科学系);Department of Atmospheric & Oceanic Science University of Maryland(马里兰大学大气与海洋科学系);Santa Fe Institute(圣菲研究所)