More Like this

1. List and Certificate Complexities in Replicable Learning

   Dixon, Peter ; Pavan, A ; Vander_Woude, Jason ; Vinodchandran, N_V ( December 2023 , Curran Associates, Inc.)

   Oh, A ; Naumann, T ; Globerson, A ; Saenko, K ; Hardt, M ; Levine, S (Ed.)

   We investigate replicable learning algorithms. Informally a learning algorithm is replicable if the algorithm outputs the same canonical hypothesis over multiple runs with high probability, even when different runs observe a different set of samples from the unknown data distribution. In general, such a strong notion of replicability is not achievable. Thus we consider two feasible notions of replicability called {\em list replicability} and {\em certificate replicability}. Intuitively, these notions capture the degree of (non) replicability. The goal is to design learning algorithms with optimal list and certificate complexities while minimizing the sample complexity. Our contributions are the following. 1. We first study the learning task of estimating the biases of $d$ coins, up to an additive error of $\varepsilon$, by observing samples. For this task, we design a $(d+1)$-list replicable algorithm. To complement this result, we establish that the list complexity is optimal, i.e there are no learning algorithms with a list size smaller than $d+1$ for this task. We also design learning algorithms with certificate complexity $\tilde{O}(\log d)$. The sample complexity of both these algorithms is $\tilde{O}(\frac{d^2}{\varepsilon^2})$ where $\varepsilon$ is the approximation error parameter (for a constant error probability). 2. In the PAC model, we show that any hypothesis class that is learnable with $d$-nonadaptive statistical queries can be learned via a $(d+1)$-list replicable algorithm and also via a $\tilde{O}(\log d)$-certificate replicable algorithm. The sample complexity of both these algorithms is $\tilde{O}(\frac{d^2}{\nu^2})$ where $\nu$ is the approximation error of the statistical query. We also show that for the concept class \dtep, the list complexity is exactly $d+1$ with respect to the uniform distribution. To establish our upper bound results we use rounding schemes induced by geometric partitions with certain properties. We use Sperner/KKM Lemma to establish the lower bound results. 

   more » « less
2. List-Decodable Linear Regression

   Karmalkar, Sushrut ; Klivans, Adam ; Kothari, Pravesh ( December 2019 , Advances in neural information processing systems)

   null (Ed.)

   We give the first polynomial-time algorithm for robust regression in the list-decodable setting where an adversary can corrupt a greater than 1/2 fraction of examples. For any our algorithm takes as input a sample of n linear equations where (1- \alpha) n of the equations are {\em arbitrarily} chosen. It outputs a list that contains a linear function that is close to the truth. Our algorithm succeeds whenever the inliers are chosen from a \emph{certifiably} anti-concentrated distribution D. In particular, this gives an efficient algorithm to find an optimal size list when the inlier distribution is standard Gaussian. For discrete product distributions that are anti-concentrated only in \emph{regular} directions, we give an algorithm that achieves similar guarantee under the promise that the true linear function has all coordinates of the same magnitude. To complement our result, we prove that the anti-concentration assumption on the inliers is information-theoretically necessary. Our algorithm is based on a new framework for list-decodable learning that strengthens the `identifiability to algorithms' paradigm based on the sum-of-squares method. 

   more » « less
3. On High-Dimensional Graph Learning Under Total Positivity

   https://doi.org/10.1109/IEEECONF53345.2021.9723191  

   Tugnait, Jitendra K. ( October 2021 , 22021 55th Asilomar Conference on Signals, Systems, and Computers)

   We consider the problem of estimating the structure of an undirected weighted sparse graphical model of multivariate data under the assumption that the underlying distribution is multivariate totally positive of order 2, or equivalently, all partial correlations are non-negative. Total positivity holds in several applications. The problem of Gaussian graphical model learning has been widely studied without the total positivity assumption where the problem can be formulated as estimation of the sparse precision matrix that encodes conditional dependence between random variables associated with the graph nodes. An approach that imposes total positivity is to assume that the precision matrix obeys the Laplacian constraints which include constraining the off-diagonal elements of the precision matrix to be non-positive. In this paper we investigate modifications to widely used penalized log-likelihood approaches to enforce total positivity but not the Laplacian structure. An alternating direction method of multipliers (ADMM) algorithm is presented for constrained optimization under total positivity and lasso as well as adaptive lasso penalties. Numerical results based on synthetic data show that the proposed constrained adaptive lasso approach significantly outperforms existing Laplacian-based approaches, both statistical and smoothness-based non-statistical. 

   more » « less
4. Distribution-Specific Auditing for Subgroup Fairness

   https://doi.org/10.4230/LIPIcs.FORC.2024.5  

   Hsu, Daniel ; Huang, Jizhou ; Juba, Brendan ( January 2024 , Schloss Dagstuhl – Leibniz-Zentrum für Informatik)

   Rothblum, Guy N (Ed.)

   We study the problem of auditing classifiers for statistical subgroup fairness. Kearns et al. [Kearns et al., 2018] showed that the problem of auditing combinatorial subgroups fairness is as hard as agnostic learning. Essentially all work on remedying statistical measures of discrimination against subgroups assumes access to an oracle for this problem, despite the fact that no efficient algorithms are known for it. If we assume the data distribution is Gaussian, or even merely log-concave, then a recent line of work has discovered efficient agnostic learning algorithms for halfspaces. Unfortunately, the reduction of Kearns et al. was formulated in terms of weak, "distribution-free" learning, and thus did not establish a connection for families such as log-concave distributions. In this work, we give positive and negative results on auditing for Gaussian distributions: On the positive side, we present an alternative approach to leverage these advances in agnostic learning and thereby obtain the first polynomial-time approximation scheme (PTAS) for auditing nontrivial combinatorial subgroup fairness: we show how to audit statistical notions of fairness over homogeneous halfspace subgroups when the features are Gaussian. On the negative side, we find that under cryptographic assumptions, no polynomial-time algorithm can guarantee any nontrivial auditing, even under Gaussian feature distributions, for general halfspace subgroups. 

   more » « less
5. Rapid Discrete Optimization via Simulation with Gaussian Markov Random Fields

   https://doi.org/10.1287/ijoc.2020.0971  

   Semelhago, Mark ; Nelson, Barry L. ; Song, Eunhye ; Wächter, Andreas ( July 2021 , INFORMS Journal on Computing)

   Inference-based optimization via simulation, which substitutes Gaussian process (GP) learning for the structural properties exploited in mathematical programming, is a powerful paradigm that has been shown to be remarkably effective in problems of modest feasible-region size and decision-variable dimension. The limitation to “modest” problems is a result of the computational overhead and numerical challenges encountered in computing the GP conditional (posterior) distribution on each iteration. In this paper, we substantially expand the size of discrete-decision-variable optimization-via-simulation problems that can be attacked in this way by exploiting a particular GP—discrete Gaussian Markov random fields—and carefully tailored computational methods. The result is the rapid Gaussian Markov Improvement Algorithm (rGMIA), an algorithm that delivers both a global convergence guarantee and finite-sample optimality-gap inference for significantly larger problems. Between infrequent evaluations of the global conditional distribution, rGMIA applies the full power of GP learning to rapidly search smaller sets of promising feasible solutions that need not be spatially close. We carefully document the computational savings via complexity analysis and an extensive empirical study. Summary of Contribution: The broad topic of the paper is optimization via simulation, which means optimizing some performance measure of a system that may only be estimated by executing a stochastic, discrete-event simulation. Stochastic simulation is a core topic and method of operations research. The focus of this paper is on significantly speeding-up the computations underlying an existing method that is based on Gaussian process learning, where the underlying Gaussian process is a discrete Gaussian Markov Random Field. This speed-up is accomplished by employing smart computational linear algebra, state-of-the-art algorithms, and a careful divide-and-conquer evaluation strategy. Problems of significantly greater size than any other existing algorithm with similar guarantees can solve are solved as illustrations. 

   more » « less