In this paper we give several applications of Littlestone dimension. The first is to the model of Angluin and Dohrn [1], where we extend their results for learning by equivalence queries with random counterexamples. Second, we extend that model to infinite concept classes with an additional source of randomness. Third, we give improved results on the relationship of Littlestone dimension to classes with extended d -compression schemes, proving the analog of a conjecture of Floyd and Warmuth [2] for Littlestone dimension.

In this paper, we give the first analyses of the non-center-based clustering objectives of sum-of-diameters and sum-of-radii under Bilu-Linial stability. Specifically, for the sum-of-diameters problem, we give polynomial-time algorithms for instances that are 2-stable, accompanied by a matching hardness result for stability below 2. For sum-of-radii clustering, we give an analysis showing that 2-stable instances are polynomial-time solvable.

Reinforcement learning generalizes multi-armed bandit problems with additional difficulties of a longer planning horizon and unknown transition kernel. We explore a black-box reduction from discounted infinite-horizon tabular reinforcement learning to multi-armed bandits, where, specifically, an independent bandit learner is placed in each state. We show that, under ergodicity and fast mixing assumptions, any slowly changing adversarial bandit algorithm achieving optimal regret in the adversarial bandit setting can also attain optimal expected regret in infinite-horizon discounted Markov decision processes, with respect to the number of rounds $T$. Furthermore, we examine our reduction using a specific instance of the exponential-weight algorithm.

In this paper we give several applications of Littlestone dimension. The first is to the model of \cite{angluin2017power}, where we extend their results for learning by equivalence queries with random counterexamples. Second, we extend that model to infinite concept classes with an additional source of randomness. Third, we give improved results on the relationship of Littlestone dimension to classes with extended $d$-compression schemes, proving a strong version of a conjecture of \cite{floyd1995sample} for Littlestone dimension.

Algorithms for noiseless collaborative PAC learning have been analyzed and optimized in recent years with respect to sample complexity. In this paper, we study collaborative PAC learning with the goal of reducing communication cost at essentially no penalty to the sample complexity. We develop communication efficient collaborative PAC learning algorithms using distributed boosting. We then consider the communication cost of collaborative learning in the presence of classification noise. As an intermediate step, we show how collaborative PAC learning algorithms can be adapted to handle classification noise. With this insight, we develop communication efficient algorithms for collaborative PAC learning robust to classification noise.

In this work, we study how to use sampling to speed up mechanisms for answering adaptive queries into datasets without reducing the accuracy of those mechanisms. This is important to do when both the datasets and the number of queries asked are very large. In particular, we describe a mechanism that provides a polynomial speed-up per query over previous mechanisms, without needing to increase the total amount of data required to maintain the same generalization error as before. We prove that this speed-up holds for arbitrary statistical queries. We also provide an even faster method for achieving statistically-meaningful responses wherein the mechanism is only allowed to see a constant number of samples from the data per query. Finally, we show that our general results yield a simple, fast, and unified approach for adaptively optimizing convex and strongly convex functions over a dataset.

In the problem of learning a class ratio from unlabeled data, which we call CR learning, the training data is unlabeled, and only the ratios, or proportions, of examples receiving each label are given. The goal is to learn a hypothesis that predicts the proportions of labels on the distribution underlying the sample. This model of learning is applicable to a wide variety of settings, including predicting the number of votes for candidates in political elections from polls. In this paper, we formally define this class and resolve foundational questions regarding the computational complexity of CR learning and characterize its relationship to PAC learning. Among our results, we show, perhaps surprisingly, that for finite VC classes what can be efficiently CR learned is a strict subset of what can be learned efficiently in PAC, under standard complexity assumptions. We also show that there exist classes of functions whose CR learnability is independent of ZFC, the standard set theoretic axioms. This implies that CR learning cannot be easily characterized (like PAC by VC dimension).

In this paper, we analyze PAC learnability from labels produced by crowdsourcing. In our setting, unlabeled examples are drawn from a distribution and labels are crowdsourced from workers who operate under classification noise, each with their own noise parameter. We develop an end-to-end crowdsourced PAC learning algorithm that takes unlabeled data points as input and outputs a trained classifier. Our three-step algorithm incorporates majority voting, pure-exploration bandits, and noisy-PAC learning. We prove several guarantees on the number of tasks labeled by workers for PAC learning in this setting and show that our algorithm improves upon the baseline by reducing the total number of tasks given to workers. We demonstrate the robustness of our algorithm by exploring its application to additional realistic crowdsourcing settings.

Learning automata is a foundational problem in computational learning theory. However, even efficiently learning random DFAs is hard. A natural restriction of this problem is to consider learning random DFAs under the uniform distribution. To date, this problem has no non-trivial lower bounds nor algorithms faster than brute force. In this note, we propose a method to find faster algorithms for this problem. We reduce the learning problem to a conjecture about meeting times of random walks over random DFAs, which may be of independent interest to prove.

In this paper, we study the problem of constructing a network by observing ordered connectivity constraints, which we define herein. These ordered constraints are made to capture realistic properties of real-world problems that are not reflected in previous, more general models. We give hardness of approximation results and nearly-matching upper bounds for the offline problem, and we study the online problem in both general graphs and restricted sub-classes. In the online problem, for general graphs, we give exponentially better upper bounds than exist for algorithms for general connectivity problems. For the restricted classes of stars and paths we are able to find algorithms with optimal competitive ratios, the latter of which involve analysis using a potential function defined over PQ-trees.

In the problem of learning with label proportions (also known as the problem of estimating class ratios), the training data is unlabeled, and only the proportions of examples receiving each label are given. The goal is to learn a hypothesis that predicts the proportions of labels on the distribution underlying the sample. This model of learning is useful in a wide variety of settings, including predicting the number of votes for candidates in political elections from polls. In this paper, we resolve foundational questions regarding the computational complexity of learning in this setting. We formalize a simple version of the setting, and we compare the computational complexity of learning in this model to classical PAC learning. Perhaps surprisingly, we show that what can be learned efficiently in this model is a strict subset of what may be leaned efficiently in PAC, under standard complexity assumptions. We give a characterization in terms of VC dimension, and we show that there are non-trivial problems in this model that can be efficiently learned. We also give an algorithm that demonstrates the feasibility of learning under well-behaved distributions.

We introduce a framework for proving lower bounds on computational problems over distributions against algorithms that can be implemented using access to a statistical query oracle. For such algorithms, access to the input distribution is limited to obtaining an estimate of the expectation of any given function on a sample drawn randomly from the input distribution rather than directly accessing samples. Most natural algorithms of interest in theory and in practice, for example, moments-based methods, local search, standard iterative methods for convex optimization, MCMC, and simulated annealing, can be implemented in this framework. Our framework is based on, and generalizes, the statistical query model in learning theory [Kearns 1998]. Our main application is a nearly optimal lower bound on the complexity of any statistical query algorithm for detecting planted bipartite clique distributions (or planted dense subgraph distributions) when the planted clique has size O ( n 1/2 − δ ) for any constant δ > 0. The assumed hardness of variants of these problems has been used to prove hardness of several other problems and as a guarantee for security in cryptographic applications. Our lower bounds provide concrete evidence of hardness, thus supporting these assumptions.

Existing approaches to active learning are generally optimistic about their certainty with respect to data shift between labeled and unlabeled data. They assume that unknown datapoint labels follow the inductive biases of the active learner. As a result, the most useful datapoint labels—ones that refute current inductive biases—are rarely solicited. We propose a shift-pessimistic approach to active learning that assumes the worst-case about the unknown conditional label distribution. This closely aligns model uncertainty with generalization error, enabling more useful label solicitation. We investigate the theoretical benefits of this approach and demonstrate its empirical advantages on probabilistic binary classification tasks.

We pose an open problem on the complexity of learning the behavior of a quantum circuit with value injection queries. We define the learning model for quantum circuits and give preliminary results. Using the test-path lemma of Angluin et al. (2009a), we show that new ideas are likely needed to tackle value injection queries for the quantum setting.

While boosting has been extensively studied, considerablyless attention has been devoted to the task of designing good weaklearning algorithms. In this paper we consider the problem of designing weak learners thatare especially adept to the boosting procedure and specifically the AdaBoost algorithm. First we describe conditions desirable for a weak learning algorithm. We then propose using sparse parity functions as weak learners, which have many of our desired properties, as weak learners in boosting. Our experimental tests show the proposed weak learners tobe competitive with the most widely used ones: decisionstumps and pruned decision trees.

We introduce a framework for proving lower bounds on computational problems over distributions, based on a class of algorithms called statistical algorithms. For such algorithms, access to the input distribution is limited to obtaining an estimate of the expectation of any given function on a sample drawn randomly from the input distribution, rather than directly accessing samples. Most natural algorithms of interest in theory and in practice, e.g., moments-based methods, local search, standard iterative methods for convex optimization, MCMC and simulated annealing, are statistical algorithms or have statistical counterparts. Our framework is inspired by and generalize the statistical query model in learning theory [34]. Our main application is a nearly optimal lower bound on the complexity of any statistical algorithm for detecting planted bipartite clique distributions (or planted dense subgraph distributions) when the planted clique has size O(n1/2-δ) for any constant δ > 0. Variants of these problems have been assumed to be hard to prove hardness for other problems and for cryptographic applications. Our lower bounds provide concrete evidence of hardness, thus supporting these assumptions.

We consider bandit problems, motivated by applications in online advertising and news story selection, in which the learner must repeatedly select a slate, that is, a subset of size s from K possible actions, and then receives rewards for just the selected actions. The goal is to minimize the regret with respect to total reward of the best slate computed in hindsight. We consider unordered and ordered versions of the problem, and give efficient algorithms which have regret O(sqrt(T)), where the constant depends on the specific nature of the problem. We also consider versions of the problem where we have access to a number of policies which make recommendations for slates in every round, and give algorithms with O(sqrt(T)) regret for competing with the best such policy as well. We make use of the technique of relative entropy projections combined with the usual multiplicative weight update algorithm to obtain our algorithms.

Boosting methods are known not to usually overfit training data even as the size of the generated classifiers becomes large. Schapire et al. attempted to explain this phenomenon in terms of the margins the classifier achieves on training examples. Later, however, Breiman cast serious doubt on this explanation by introducing a boosting algorithm, arc-gv, that can generate a higher margins distribution than AdaBoost and yet performs worse. In this paper, we take a close look at Breiman’s compelling but puzzling results. Although we can reproduce his main finding, we find that the poorer performance of arc-gv can be explained by the increased complexity of the base classifiers it uses, an explanation supported by our experiments and entirely consistent with the margins theory. Thus, we find maximizing the margins is desirable, but not necessarily at the expense of other factors, especially baseclassifier complexity. 1.