arXiv research

Tutorial on using concentration inequalities for linear system identification.

New sampling method optimizes learning minimum mean among distributions.

Study improves self-normalized bounds for vector-valued processes beyond sub-Gaussianity.

Calculation of the log-normalizer is a major computational obstacle in applications of log-linear models with large output spaces. The problem of fast normalizer computation has therefore attracted significant attention in the theoretical and applied machine learning literature. In this paper, we analyze a recently pro…

New bounds on self-normalized martingales improve online linear regression performance.

We improve bounds for stochastic processes, especially those with heavy tails.

A new method for evaluating and selecting policies in contextual bandits improves confidence intervals and policy quality.

We study the regret minimization problem in the novel setting of generalized kernelized bandits (GKBs), where we optimize an unknown function $f^*$ belonging to a reproducing kernel Hilbert space (RKHS) having access to samples generated by an exponential family (EF) reward model whose mean is a non-linear function $μ(…

Unified stopping rules ensure accurate policies in contextual learning.

Deviation inequalities for stochastic approximation methods.

Deviation inequalities and limit laws for random walks on metric spaces.

New deviation inequalities for multi-armed bandit models.

New inequalities for matrix supermartingales converge under various conditions.

New inequality criterion for a mean field equation on spheres.

New Gini indices capture more nuanced income inequality.

We study random walks on groups with the feature that, roughly speaking, successive positions of the walk tend to be "aligned". We formalize and quantify this property by means of the notion of deviation inequalities. We show that deviation inequalities have several consequences including Central Limit Theorems, the lo…

The paper improves inequalities for nearly spherical sets using quermassintegrals.

Novel concentration inequalities are obtained for the missing mass, i.e. the total probability mass of the outcomes not observed in the sample. We derive distribution-free deviation bounds with sublinear exponents in deviation size for missing mass and improve the results of Berend and Kontorovich (2013) and Yari Saeed…

The paper provides a finite-sample deviation bound for stable autoregressive processes.

In this paper, we study the risk bounds for samples independently drawn from an infinitely divisible (ID) distribution. In particular, based on a martingale method, we develop two deviation inequalities for a sequence of random variables of an ID distribution with zero Gaussian component. By applying the deviation ineq…

Refines online learning to rank algorithm with tighter bounds.

BR-SNIS reduces bias in self-normalized IS without increasing variance.

New algorithm reduces regret for logistic bandits without $κ$ dependency.

The stochastic multi-armed bandit model is a simple abstraction that has proven useful in many different contexts in statistics and machine learning. Whereas the achievable limit in terms of regret minimization is now well known, our aim is to contribute to a better understanding of the performance in terms of identify…

Proves inequality linking function deviation to gradient norm on compact manifolds.

We are concerned with obtaining novel concentration inequalities for the missing mass, i.e. the total probability mass of the outcomes not observed in the sample. We not only derive - for the first time - distribution-free Bernstein-like deviation bounds with sublinear exponents in deviation size for missing mass, but …

Improved speech recognition with faster training and inference.

A tutorial on non-asymptotic system identification methods.

The paper analyzes stress shocks and their impact on historical data.

The paper develops a method for self-normalized inference in adaptive experiments.

New algorithms reduce regret in neural logistic bandits.

Sharp inequalities for matrix means with unknown variance.

Sharp concentration results for sums of heavy-tailed random variables.

This paper studies semiparametric contextual bandits, a generalization of the linear stochastic bandit problem where the reward for an action is modeled as a linear function of known action features confounded by an non-linear action-independent term. We design new algorithms that achieve $\tilde{O}(d\sqrt{T})$ regret …

The paper provides bounds for high-dimensional U-statistics with novel order-explicit inequalities.

Study on discrepancy principle for learning algorithms in nonparametric regression.

A new linear contextual bandit algorithm with improved regret bound.

Sharp concentration bounds for i.i.d. variables.

New inequality on sphere generalizes circle inequality.

Self Normalizing Flows improve normalizing flows by reducing computational complexity.

In this paper, we are concerned with obtaining distribution-free concentration inequalities for mixture of independent Bernoulli variables that incorporate a notion of variance. Missing mass is the total probability mass associated to the outcomes that have not been seen in a given sample which is an important quantity…

Improved concentration inequalities for sub-Weibull variables enhance statistical and machine learning applications.

The paper improves generalization bounds for classifier chains with interdependent labels.

Paper develops a new inequality for non-causal machine learning.

New activation function SERLU improves neural network performance.

Sharp upper bounds derived for Alexandrov-Fenchel deficit using weighted Minkowski integral formulas.

This article provides a new toolbox to derive sparse recovery guarantees from small deviations on extreme singular values or extreme eigenvalues obtained in Random Matrix Theory. This work is based on Restricted Isometry Constants (RICs) which are a pivotal notion in Compressed Sensing and High-Dimensional Statistics a…

Estimates stationary mass and frequency from non-i.i.d. data.