Unknown Intervention

Research on unknown interventions focuses on inferring causal relationships from data generated under interventions where the specific variables manipulated are not known. Current approaches leverage Bayesian methods, structural causal models, and novel algorithms to identify intervention targets and reconstruct causal graphs, often employing non-parametric techniques to handle complex, high-dimensional data and avoid restrictive assumptions like linearity. This work is significant for advancing causal discovery in scenarios with limited knowledge of experimental manipulations, impacting fields like biology and medicine where interventions are often imperfectly controlled or their effects are indirect and complex. Improved methods for handling unknown interventions will enable more robust and reliable causal inference from real-world data.

Papers

March 27, 2024

Non-Linear Inference Time Intervention: Improving LLM Truthfulness
Jakub Hoscilowicz, Adam Wiacek, Jan Chojnacki, Adam Cieslak, Leszek Michon, Vitalii Urbanevych, Artur Janicki
Representation Space Multiple Choice LLM Truthfulness Unknown Intervention Inference Time Intervention

December 11, 2023

Learning Unknown Intervention Targets in Structural Causal Models from Heterogeneous Data
Yuqin Yang, Saber Salehkaleybar, Negar Kiyavash
LeArning Abstract Causal Model Structural Causal Model Heterogeneous Data Latent Confounders Global Endogenous Variate Unknown Intervention Causal Overhypotheses

December 1, 2023

Bayesian causal discovery from unknown general interventions
Alessandro Mascaro, Federico Castelletti
Causal Discovery Bayesian Inference Directed Acyclic Graph Markov Equivalence Class Interventional Data Bayesian Causal Discovery Unknown Intervention

June 5, 2023

Learning nonparametric latent causal graphs with unknown interventions
Yibo Jiang, Bryon Aragam
Latent Space Latent Causal Latent Structure Parametric Model Non Parametric Learning Unknown Intervention

June 1, 2023

Nonparametric Identifiability of Causal Representations from Unknown Interventions
Julius von Kügelgen, Michel Besserve, Liang Wendong, Luigi Gresele, Armin Kekić, Elias Bareinboim, David M. Blei, Bernhard Schölkopf
Causal Representation Learning Latent Causal Causal Representation Underlying Causal Unknown Intervention

October 1, 2022

Disentangling Mixtures of Unknown Causal Interventions
Abhinav Kumar, Gaurav Sinha
Partial Identifiability Learning Mixture Mixture Weight Causal Bayesian Network Genetic Perturbation Node Intervention Unknown Intervention

June 3, 2022

BaCaDI: Bayesian Causal Discovery with Unknown Interventions
Alexander Hägele, Jonas Rothfuss, Lars Lorch, Vignesh Ram Somnath, Bernhard Schölkopf, Andreas Krause
Causal Structure Causal Discovery Method Unknown Intervention Synthetic Causal