Sample efficient active learning of causal trees
Name
NeurIPS-2019-sample-efficient-active-learning-of-causal-trees-Paper.pdf
Description
Published version
Size
824.68 KB
Format
Adobe PDF
Checksum (MD5)
8a62e84478d99ab6c9effe9c1aa58383
Author(s)
Greenewald, Kristjan
Katz, Dmitriy
Shanmugam, Karthikeyan
Magliacane, Sara
Kocaoglu, Murat
Boix-Adsera, Enric
Bresler, Guy
Date Issued
December 2019
Journal
Advances in Neural Information Processing Systems
Publisher
Morgan Kaufmann Publishers
Citation
Greenewald, Kristjan et al. “Sample efficient active learning of causal trees.” Paper in the Advances in Neural Information Processing Systems, 32, 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancourver Canada, Dec 8-14, 2019, Morgan Kaufmann © 2019 The Author(s)
Version
Final published version
Abstract
Neural information processing systems foundation. All rights reserved. We consider the problem of experimental design for learning causal graphs that have a tree structure. We propose an adaptive framework that determines the next intervention based on a Bayesian prior updated with the outcomes of previous experiments, focusing on the setting where observational data is cheap (assumed infinite) and interventional data is expensive. While information greedy approaches are popular in active learning, we show that in this setting they can be exponentially suboptimal (in the number of interventions required), and instead propose an algorithm that exploits graph structure in the form of a centrality measure. If each intervention yields a very large data sample, we show that the algorithm requires a number of interventions less than or equal to a factor of 2 times the minimum achievable number. We show that the algorithm and the associated theory can be adapted to the setting where each performed intervention yields finitely many samples. Several extensions are also presented, to the case where a specified set of nodes cannot be intervened on, to the case where K interventions are scheduled at once, and to the fully adaptive case where each experiment yields only one sample. In the case of finite interventional data, through simulated experiments we show that our algorithms outperform different adaptive baseline algorithms.
MIT Department
MIT-IBM Watson AI Lab
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://papers.nips.cc/paper/2019/hash/5ee5605917626676f6a285fa4c10f7b0-Abstract.html
