arXiv research

Three new efficient algorithms project vectors onto weighted l1 ball.

GPE algorithm optimizes nonparametric contextual bandits with efficient regret bounds.

Efficient algorithms solve joint graphical lasso problems.

Efficient algorithm reduces communication costs in sparse regression.

NNs can learn efficient algorithms for certain problems.

Incremental versions of batch algorithms are often desired, for increased time efficiency in the streaming data setting, or increased memory efficiency in general. In this paper we present a novel algorithm for incremental kernel PCA, based on rank one updates to the eigendecomposition of the kernel matrix, which is mo…

We describe a framework for designing efficient active learning algorithms that are tolerant to random classification noise and are differentially-private. The framework is based on active learning algorithms that are statistical in the sense that they rely on estimates of expectations of functions of filtered random e…

Efficient WKNN-Shapley computation improves data valuation accuracy.

We introduce a novel and efficient sampling algorithm for the Multiplicative Attribute Graph Model (MAGM - Kim and Leskovec (2010)}). Our algorithm is \emph{strictly} more efficient than the algorithm proposed by Yun and Vishwanathan (2012), in the sense that our method extends the \emph{best} time complexity guarantee…

New algorithm for online omniprediction with strong guarantees for continuous hypothesis classes.

New algorithms improve policy evaluation in reinforcement learning.

Efficient algorithms for clustered Lasso and OSCAR reduce computational costs.

Study efficient auditing of ML fairness models.

New algorithm estimates treatment effects for more efficient contextual bandits.

Unified framework for efficient Frank-Wolfe optimization of Dominant Set Clustering.

Efficient algorithm for halfspaces with specific noise conditions.

Memory-efficient algorithm reduces variance in off-policy RL.

This paper introduces STAP to measure DEX efficiency and shows better routing algorithms increase DEX performance and stakeholder benefits.

Efficiently factorizes coupled matrix tensor data for better accuracy and speed.

Efficient algorithms for contextual bandits with smooth regret in continuous action spaces.

AI progress measured by reduced compute needed to reach past performance.

Communication-efficient SGD algorithms, which allow nodes to perform local updates and periodically synchronize local models, are highly effective in improving the speed and scalability of distributed SGD. However, a rigorous convergence analysis and comparative study of different communication-reduction strategies rem…

Efficient algorithm solves best subset selection problem.

Improves reinforcement learning stability and efficiency.

Matrix profile has been recently proposed as a promising technique to the problem of all-pairs-similarity search on time series. Efficient algorithms have been proposed for computing it, e.g., STAMP, STOMP and SCRIMP++. All these algorithms use the z-normalized Euclidean distance to measure the distance between subsequ…

New algorithm finds near-optimal policies efficiently in zero-sum games.

Study efficient learning of robust halfspaces with noise.

Model-free Reinforcement Learning (RL) algorithms such as Q-learning [Watkins, Dayan 92] have been widely used in practice and can achieve human level performance in applications such as video games [Mnih et al. 15]. Recently, equipped with the idea of optimism in the face of uncertainty, Q-learning algorithms [Jin, Al…

Efficient algorithm for mobile health provides timely physical activity suggestions.

Practical algorithm for contextual bandits with large action spaces.

Three new oracle-efficient algorithms for private synthetic data release.

Survey of robust statistical methods for efficient computation.

Some methods based on simple regularizing geometric element transformations have heuristically been shown to give runtime efficient and quality effective smoothing algorithms for meshes. We describe the mathematical framework and a systematic approach to global optimization-based versions of such methods for mixed volu…

This paper addresses the problem of neighborhood selection for Gaussian graphical models. We present two heuristic algorithms: a forward-backward greedy algorithm for general Gaussian graphical models based on mutual information test, and a threshold-based algorithm for walk summable Gaussian graphical models. Both alg…

Efficient streaming algorithms for robust statistics with near-optimal memory.

New algorithms reduce reinforcement learning regret in factored MDPs.

New RL algorithms improve control tasks with data reuse.

We propose an efficient algorithm for sparse signal reconstruction problems. The proposed algorithm is an augmented Lagrangian method based on the dual sparse reconstruction problem. It is efficient when the number of unknown variables is much larger than the number of observations because of the dual formulation. More…

Efficient strategies for online learning against bandit algorithms solve minimax problems.

We study the problem of efficient online multiclass linear classification with bandit feedback, where all examples belong to one of $K$ classes and lie in the $d$-dimensional Euclidean space. Previous works have left open the challenge of designing efficient algorithms with finite mistake bounds when the data is linear…

Paper tackles sample-efficient offline RL, proposing data diversity and unified algorithms.

Efficiently private clustering algorithms with tight approximation ratios.

NPOD algorithm improves efficiency in estimating pharmacokinetic parameters.

We introduce GLR-klUCB, a novel algorithm for the piecewise iid non-stationary bandit problem with bounded rewards. This algorithm combines an efficient bandit algorithm, kl-UCB, with an efficient, parameter-free, changepoint detector, the Bernoulli Generalized Likelihood Ratio Test, for which we provide new theoretica…

Oracle-efficient algorithms for online learning with smoothed and hint-adversaries.

Paper establishes NE existence and efficient algorithms for weakly monotone GMFGs.

New efficient algorithm for approximate PML distribution.

Learning in the presence of outliers is a fundamental problem in statistics. Until recently, all known efficient unsupervised learning algorithms were very sensitive to outliers in high dimensions. In particular, even for the task of robust mean estimation under natural distributional assumptions, no efficient algorith…