More Like this

1. Learning and Testing Causal Models with Interventions

   Acharya, Jayadev; Bhattacharyya, Arnab; Daskalakis, Constantinos; Kandasamy, Saravanan (January 2018, 32nd Annual Conference on Neural Information Processing Systems)

   We consider testing and learning problems on causal Bayesian networks as defined by Pearl (Pearl, 2009). Given a causal Bayesian network  on a graph with n discrete variables and bounded in-degree and bounded `confounded components', we show that O(logn) interventions on an unknown causal Bayesian network  on the same graph, and Õ (n/ϵ2) samples per intervention, suffice to efficiently distinguish whether = or whether there exists some intervention under which  and  are farther than ϵ 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: Ω(logn) 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. 

   more » « less
2. Square Hellinger Subadditivity for Bayesian Networks and its Applications to Identity Testing.

   Daskalakis, Constantinos; Pan, Qinxuan (January 2017, 30th Annual Conference on Learning Theory)

   We show that the square Hellinger distance between two Bayesian networks on the same directed graph, G, is subadditive with respect to the neighborhoods of G. Namely, if P and Q are the probability distributions defined by two Bayesian networks on the same DAG, our inequality states that the square Hellinger distance, H2(P,Q), between P and Q is upper bounded by the sum, ∑vH2(P{v}∪Πv,Q{v}∪Πv), of the square Hellinger distances between the marginals of P and Q on every node v and its parents Πv in the DAG. Importantly, our bound does not involve the conditionals but the marginals of P and Q. We derive a similar inequality for more general Markov Random Fields. As an application of our inequality, we show that distinguishing whether two Bayesian networks P and Q on the same (but potentially unknown) DAG satisfy P=Q vs dTV(P,Q)>ϵ can be performed from Õ (|Σ|3/4(d+1)⋅n/ϵ2) samples, where d is the maximum in-degree of the DAG and Σ the domain of each variable of the Bayesian networks. If P and Q are defined on potentially different and potentially unknown trees, the sample complexity becomes Õ (|Σ|4.5n/ϵ2), whose dependence on n,ϵ is optimal up to logarithmic factors. Lastly, if P and Q are product distributions over {0,1}n and Q is known, the sample complexity becomes O(n‾√/ϵ2), which is optimal up to constant factors. 

   more » « less
3. Learning and Sampling of Atomic Interventions from Observations

   Bhattacharyya, Arnab; Gayen, Sutanu; Kandasamy, Saravanan; Maran, Ashwin; Vinodchandran, N. V. (July 2020, Proceedings of the 37th International Conference on Machine Learning)

   We study the problem of efficiently estimating the effect of an intervention on a single variable using observational samples. Our goal is to give algorithms with polynomial time and sample complexity in a non-parametric setting. Tian and Pearl (AAAI ’02) have exactly characterized the class of causal graphs for which causal effects of atomic interventions can be identified from observational data. We make their result quantitative. Suppose 𝒫 is a causal model on a set V of n observable variables with respect to a given causal graph G, and let do(x) be an identifiable intervention on a variable X. We show that assuming that G has bounded in-degree and bounded c-components (k) and that the observational distribution satisfies a strong positivity condition: (i) [Evaluation] There is an algorithm that outputs with probability 2/3 an evaluator for a distribution P^ that satisfies TV(P(V | do(x)), P^(V)) < eps using m=O (n/eps^2) samples from P and O(mn) time. The evaluator can return in O(n) time the probability P^(v) for any assignment v to V. (ii) [Sampling] There is an algorithm that outputs with probability 2/3 a sampler for a distribution P^ that satisfies TV(P(V | do(x)), P^(V)) < eps using m=O (n/eps^2) samples from P and O(mn) time. The sampler returns an iid sample from P^ with probability 1 in O(n) time. We extend our techniques to estimate P(Y | do(x)) for a subset Y of variables of interest. We also show lower bounds for the sample complexity, demonstrating that our sample complexity has optimal dependence on the parameters n and eps, as well as if k=1 on the strong positivity parameter. 

   more » « less
4. Private High-Dimensional Hypothesis Testing

   Narayanan, S. (January 2022, Conference on Learning Theory (COLT 2022))

   We provide improved differentially private algorithms for identity testing of high-dimensional distributions. Specifically, for d-dimensional Gaussian distributions with known covariance Σ, we can test whether the distribution comes from N(μ∗,Σ) for some fixed μ∗ or from some N(μ,Σ) with total variation distance at least α from N(μ∗,Σ) with (ε,0)-differential privacy, using only O~(d1/2α2+d1/3α4/3⋅ε2/3+1α⋅ε) samples if the algorithm is allowed to be computationally inefficient, and only O~(d1/2α2+d1/4α⋅ε) samples for a computationally efficient algorithm. We also provide a matching lower bound showing that our computationally inefficient algorithm has optimal sample complexity. We also extend our algorithms to various related problems, including mean testing of Gaussians with bounded but unknown covariance, uniformity testing of product distributions over {−1,1}d, and tolerant testing. Our results improve over the previous best work of Canonne et al.~\cite{CanonneKMUZ20} for both computationally efficient and inefficient algorithms, and even our computationally efficient algorithm matches the optimal \emph{non-private} sample complexity of O(d√α2) in many standard parameter settings. In addition, our results show that, surprisingly, private identity testing of d-dimensional Gaussians can be done with fewer samples than private identity testing of discrete distributions over a domain of size d \cite{AcharyaSZ18}, which refutes a conjectured lower bound of~\cite{CanonneKMUZ20}. 

   more » « less
5. Lower bounds for testing graphical models: colorings and antiferromagnetic Ising models

   Bezakova, Ivona; Blanca, Antonio; Chen, Zongchen; Stefankovic, Daniel; Vigoda, Eric (July 2019, Proceedings of Machine Learning Research)

   We study the identity testing problem in the context of spin systems or undirected graphical models, where it takes the following form: given the parameter specification of the model M and a sampling oracle for the distribution \mu_{M^*} of an unknown model M^*, can we efficiently determine if the two models M and M^* are the same? We consider identity testing for both soft-constraint and hard-constraint systems. In particular, we prove hardness results in two prototypical cases, the Ising model and proper colorings, and explore whether identity testing is any easier than structure learning. For the ferromagnetic (attractive) Ising model, Daskalasis et al. (2018) presented a polynomial time algorithm for identity testing. We prove hardness results in the antiferromagnetic (repulsive) setting in the same regime of parameters where structure learning is known to require a super-polynomial number of samples. In particular, for n-vertex graphs of maximum degree d, we prove that if |\beta| d = \omega(\log n) (where \beta is the inverse temperature parameter), then there is no identity testing algorithm for the antiferromagnetic Ising model that runs in polynomial time unless RP = NP. We also establish computational lower bounds for a broader set of parameters under the (randomized) exponential time hypothesis. In our proofs, we use random graphs as gadgets; this is inspired by similar constructions in seminal works on the hardness of approximate counting. In the hard-constraint setting, we present hardness results for identity testing for proper colorings. Our results are based on the presumed hardness of #BIS, the problem of (approximately) counting independent sets in bipartite graphs. In particular, we prove that identity testing for colorings is hard in the same range of parameters where structure learning is known to be hard, which in turn matches the parameter regime for NP-hardness of the corresponding decision problem. 

   more » « less