Acharya, Jayadev and Bhattacharyya, Arnab and Daskalakis, Constantinos and Kandasamy, Saravanan (2018) Learning and Testing Causal Models with Interventions. In: 32nd Conference on Neural Information Processing Systems (NIPS), December 2 - 8, 2018, Montreal, Canada.
|
PDF
Adv_Neu_Inf_Pro_Sys_31_2018.pdf - Published Version Download (399kB) | Preview |
Abstract
We consider testing and learning problems on causal Bayesian networks as defined by Pearl Pea09]. Given a causal Bayesian network M on a graph with n discrete variables and bounded in-degree and bounded ``confounded components'', we show that O(log n) interventions on an unknown causal Bayesian network X on the same graph, and O(n/epsilon(2)) samples per intervention, suffice to efficiently distinguish whether X = M or whether there exists some intervention under which X and M are farther than epsilon in total variation distance. We also obtain sample/time/intervention efficient algorithms for: (i) testing the identity of two unknown causal Bayesian networks on the same graph; and (ii) learning a causal Bayesian network on a given graph. Although our algorithms are non-adaptive, we show that adaptivity does not help in general: Omega(log n) interventions are necessary for testing the identity of two unknown causal Bayesian networks on the same graph, even adaptively. Our algorithms are enabled by a new subadditivity inequality for the squared Hellinger distance between two causal Bayesian networks.
| Item Type: | Conference Proceedings |
|---|---|
| Series.: | Advances in Neural Information Processing Systems |
| Publisher: | NEURAL INFORMATION PROCESSING SYSTEMS (NIPS) |
| Additional Information: | Copyright belongs to NEURAL INFORMATION PROCESSING SYSTEMS (NIPS) |
| Keywords: | Distributions;Selection;Circuits |
| Department/Centre: | Division of Electrical Sciences > Computer Science & Automation |
| Date Deposited: | 21 Apr 2019 08:06 |
| Last Modified: | 21 Apr 2019 08:06 |
| URI: | http://eprints.iisc.ac.in/id/eprint/62308 |
Actions (login required)
 |
View Item |
