Search Results for author:
Found 34 papers, 9 papers with code
Bounding the fairness and accuracy of classifiers from population statistics
We consider the study of a classification model whose properties are impossible to estimate using a validation set, either due to the absence of such a set or because access to the classifier, even as a black-box, is impossible.
Adaptive Combinatorial Maximization: Beyond Approximate Greedy Policies
Our approximation guarantees simultaneously support the maximal gain ratio as well as near-submodular utility functions, and include both maximization under a cardinality constraint and a minimum cost coverage guarantee.
On the Capacity Limits of Privileged ERM
We provide a counter example to a claim made in that work regarding the VC dimension of the loss class induced by this problem; We conclude that the claim is incorrect.
Improved Robust Algorithms for Learning with Discriminative Feature Feedback
In this work, we provide new robust interactive learning algorithms for the Discriminative Feature Feedback model, with mistake bounds that are significantly lower than those of previous robust algorithms for this setting.
Fairness and Unfairness in Binary and Multiclass Classification: Quantifying, Calculating, and Bounding
We propose a new interpretable measure of unfairness, that allows providing a quantitative analysis of classifier fairness, beyond a dichotomous fair/unfair distinction.
Fast Distributed k-Means with a Small Number of Rounds
Our algorithm guarantees a cost approximation factor and a number of communication rounds that depend only on the computational capacity of the coordinator.
A Fast Algorithm for PAC Combinatorial Pure Exploration
We provide sample complexity guarantees for our algorithm, and demonstrate in experiments its usefulness on large problems, whereas previous algorithms are impractical to run on problems of even a few dozen arms.
Active Structure Learning of Bayesian Networks in an Observational Setting
We show that for a class of distributions that we term stable, a sample complexity reduction of up to a factor of $\widetilde{\Omega}(d^3)$ can be obtained, where $d$ is the number of network variables.
A Constant Approximation Algorithm for Sequential Random-Order No-Substitution k-Median Clustering
We give the first algorithm for this setting that obtains a constant approximation factor on the optimal risk under a random arrival order, an exponential improvement over previous work.
Active Feature Selection for the Mutual Information Criterion
We study active feature selection, a novel feature selection setting in which unlabeled data is available, but the budget for labels is limited, and the examples to label can be actively selected by the algorithm.
Approximating a Target Distribution using Weight Queries
The algorithm finds a reweighting of the data set that approximates the weights according to the target distribution, using a limited number of weight queries.
Robust Learning from Discriminative Feature Feedback
We show how such errors can be handled algorithmically, in both an adversarial and a stochastic setting.
Epsilon-Best-Arm Identification in Pay-Per-Reward Multi-Armed Bandits
We study epsilon-best-arm identification, in a setting where during the exploration phase, the cost of each arm pull is proportional to the expected future reward of that arm.
Universal Bayes consistency in metric spaces
This is the first learning algorithm known to enjoy this property; by comparison, the $k$-NN classifier and its variants are not generally universally Bayes-consistent, except under additional structural assumptions, such as an inner product, a norm, finite dimension, or a Besicovitch-type property.
Sequential no-Substitution k-Median-Clustering
We provide an efficient algorithm for this setting, and show that its multiplicative approximation factor is twice the approximation factor of an efficient offline algorithm.
Learning from discriminative feature feedback
We consider the problem of learning a multi-class classifier from labels as well as simple explanations that we call "discriminative features".
The Principle of Logit Separation
In contrast, there are known loss functions, as well as novel batch loss functions that we propose, which are aligned with this principle.
Temporal anomaly detection: calibrating the surprise
We learn a low-rank stationary model from the training data, and then fit a regression model for predicting the expected likelihood score of normal access patterns in the future.
Fast Single-Class Classification and the Principle of Logit Separation
Our experiments show that indeed in almost all cases, losses that are aligned with the Principle of Logit Separation obtain at least 20% relative accuracy improvement in the SLC task compared to losses that are not aligned with it, and sometimes considerably more.
Nearest-Neighbor Sample Compression: Efficiency, Consistency, Infinite Dimensions
We examine the Bayes-consistency of a recently proposed 1-nearest-neighbor-based multiclass learning algorithm.
Tunable Sensitivity to Large Errors in Neural Network Training
We propose incorporating this idea of tunable sensitivity for hard examples in neural network learning, using a new generalization of the cross-entropy gradient step, which can be used in place of the gradient in any gradient-based training method.
Active Nearest-Neighbor Learning in Metric Spaces
We propose a pool-based non-parametric active learning algorithm for general metric spaces, called MArgin Regularized Metric Active Nearest Neighbor (MARMANN), which outputs a nearest-neighbor classifier.
Submodular Learning and Covering with Response-Dependent Costs
We further show that in both settings, the approximation factor of this greedy algorithm is near-optimal among all greedy algorithms.
Interactive algorithms: from pool to stream
We consider interactive algorithms in the pool-based setting, and in the stream-based setting.
Active Regression by Stratification
We propose a new active learning algorithm for parametric linear regression with random design.
Multiclass learnability and the ERM principle
We study the sample complexity of multiclass prediction in several learning settings.
Loss minimization and parameter estimation with heavy tails
This work studies applications and generalizations of a simple estimation technique that provides exponential concentration under heavy-tailed distributions, assuming only bounded low-order moments.
Auditing: Active Learning with Outcome-Dependent Query Costs
We term the setting auditing, and consider the auditing complexity of an algorithm: the number of negative labels the algorithm requires in order to learn a hypothesis with low relative error.
Feature Multi-Selection among Subjective Features
When dealing with subjective, noisy, or otherwise nebulous features, the "wisdom of crowds" suggests that one may benefit from multiple judgments of the same feature on the same object.
Learning Sparse Low-Threshold Linear Classifiers
We consider the problem of learning a non-negative linear classifier with a $1$-norm of at most $k$, and a fixed threshold, under the hinge-loss.
Multiclass Learning Approaches: A Theoretical Comparison with Implications
We analyze both the estimation error and the approximation error of these methods.
Efficient Active Learning of Halfspaces: an Aggressive Approach
Our efficient aggressive active learner of half-spaces has formal approximation guarantees that hold when the pool is separable with a margin.
Distribution-Dependent Sample Complexity of Large Margin Learning
We obtain a tight distribution-specific characterization of the sample complexity of large-margin classification with L2 regularization: We introduce the margin-adapted dimension, which is a simple function of the second order statistics of the data distribution, and show distribution-specific upper and lower bounds on the sample complexity, both governed by the margin-adapted dimension of the data distribution.
Tight Sample Complexity of Large-Margin Learning
We obtain a tight distribution-specific characterization of the sample complexity of large-margin classification with L2 regularization: We introduce the gamma-adapted-dimension, which is a simple function of the spectrum of a distribution's covariance matrix, and show distribution-specific upper and lower bounds on the sample complexity, both governed by the gamma-adapted-dimension of the source distribution.
