arXiv research

CD algorithm achieves near-optimal convergence rate for unnormalized models.

New algorithms improve robust estimation in contaminated Gaussian models.

Near-optimal rates for multi-task learning with shared representations.

In this paper, we investigate the statistical convergence rate of a Bayesian low-rank tensor estimator. Our problem setting is the regression problem where a tensor structure underlying the data is estimated. This problem setting occurs in many practical applications, such as collaborative filtering, multi-task learnin…

Develops a high-dimensional differentially-private EM algorithm with near-optimal statistical guarantees.

New algorithm solves complex optimization problems efficiently.

Alt-GDA outperforms Sim-GDA in minimax games with near-optimal local convergence.

2D CNNs approximate Korobov functions with near-optimal rates.

In quantitative finance, we often model asset prices as semimartingales, with drift, diffusion and jump components. The jump activity index measures the strength of the jumps at high frequencies, and is of interest both in model selection and fitting, and in volatility estimation. In this paper, we give a novel estimat…

We decode latent states in Block MDPs and learn near-optimal policies.

Bayesian deep learning with heavy-tailed weights achieves near-optimal performance.

Rejection Sampling is a fundamental Monte-Carlo method. It is used to sample from distributions admitting a probability density function which can be evaluated exactly at any given point, albeit at a high computational cost. However, without proper tuning, this technique implies a high rejection rate. Several methods h…

A new UCB algorithm for heavy-tailed bandits with near-optimal regret.

New method finds near-optimal solutions for non-convex optimization problems.

Algorithm extsc{Pedel} learns near-optimal policies efficiently on specific problems.

Near-optimal private tests for simple and MLR hypotheses developed under Gaussian differential privacy.

New algorithm clusters trajectories from multiple Markov chains with near-optimal error.

Algorithm learns linear systems from partial observations with near-optimal rate.

Paper quantizes heavy-tailed data for near optimal estimation rates.

Improved Clipped-SGD achieves near-optimal heavy-tailed statistical estimation in streaming settings.

Based on the tree architecture, the objective of this paper is to design deep neural networks with two or more hidden layers (called deep nets) for realization of radial functions so as to enable rotational invariance for near-optimal function approximation in an arbitrarily high dimensional Euclidian space. It is show…

Paper develops DP methods for low-rank matrix estimation with near-optimal performance.

We consider estimating the parametric components of semi-parametric multiple index models in a high-dimensional and non-Gaussian setting. Such models form a rich class of non-linear models with applications to signal processing, machine learning and statistics. Our estimators leverage the score function based first and…

New algorithms ensure reproducibility and optimal convergence in convex optimization.

There has been a recent surge of interest in studying permutation-based models for ranking from pairwise comparison data. Despite being structurally richer and more robust than parametric ranking models, permutation-based models are less well understood statistically and generally lack efficient learning algorithms. In…

We present the first adaptive strategy for active learning in the setting of classification with smooth decision boundary. The problem of adaptivity (to unknown distributional parameters) has remained opened since the seminal work of Castro and Nowak (2007), which first established (active learning) rates for this sett…

New algorithms achieve near-optimal cumulative loss in nonparametric online learning and games.

Policy gradient algorithm with variable learning rates achieves near-optimal performance in multi-arm bandit problems.

New asymptotic e-values improve inference by eliminating data-dependent scaling inefficiency.

Stochastic gradient descent (SGD) gives an optimal convergence rate when minimizing convex stochastic objectives $f(x)$. However, in terms of making the gradients small, the original SGD does not give an optimal rate, even when $f(x)$ is convex. If $f(x)$ is convex, to find a point with gradient norm $\varepsilon$, we …

This work demonstrates the potential of deep reinforcement learning techniques for transmit power control in wireless networks. Existing techniques typically find near-optimal power allocations by solving a challenging optimization problem. Most of these algorithms are not scalable to large networks in real-world scena…

SGD achieves near optimal convergence rate in smooth interpolation regime.

We present an improved analysis of the Euler-Maruyama discretization of the Langevin diffusion. Our analysis does not require global contractivity, and yields polynomial dependence on the time horizon. Compared to existing approaches, we make an additional smoothness assumption, and improve the existing rate from $O(η)…

ROOT-SGD solves convex optimization problems with optimal nonasymptotic and near-optimal asymptotic performance.

We study pool-based active learning with abstention feedbacks where a labeler can abstain from labeling a queried example with some unknown abstention rate. This is an important problem with many useful applications. We take a Bayesian approach to the problem and develop two new greedy algorithms that learn both the cl…

Post-training optimizes model performance beyond base model limits.

New algorithms robustly estimate mean with near-optimal error rates.

ODCGM solves non-convex optimization on manifolds with simpler projections.

Large learning rates improve neural network generalization, study shows.

We study pool-based active learning with abstention feedbacks, where a labeler can abstain from labeling a queried example with some unknown abstention rate. This is an important problem with many useful applications. We take a Bayesian approach to the problem and develop two new greedy algorithms that learn both the c…

New DP optimization methods for sparse gradients, improving on existing algorithms.

Recent work on follow the perturbed leader (FTPL) algorithms for the adversarial multi-armed bandit problem has highlighted the role of the hazard rate of the distribution generating the perturbations. Assuming that the hazard rate is bounded, it is possible to provide regret analyses for a variety of FTPL algorithms f…

We solve robust regression and matrix completion problems with sparse and low-rank models.

New adaptive SGD algorithms for federated learning over physical channels.

We consider a stochastic bandit problem with infinitely many arms. In this setting, the learner has no chance of trying all the arms even once and has to dedicate its limited number of samples only to a certain number of arms. All previous algorithms for this setting were designed for minimizing the cumulative regret o…

We formulate statistical watermarking as hypothesis testing and establish near-optimal bounds.

Local update methods' performance depends on learning rates, affecting convergence rates and alignment with true loss.

Polynomial-time algorithm for near-optimal community detection in graphs.