arXiv research

New data-dependent priors improve PAC-Bayes bounds.

The aim of this paper is to provide some theoretical understanding of quasi-Bayesian aggregation methods non-negative matrix factorization. We derive an oracle inequality for an aggregated estimator. This result holds for a very general class of prior distributions and shows how the prior affects the rate of convergenc…

Oracle inequality for sparse neural nets adapts to unknown structure.

We presented Bayesian portfolio selection strategy, via the $k$ factor asset pricing model. If the market is information efficient, the proposed strategy will mimic the market; otherwise, the strategy will outperform the market. The strategy depends on the selection of a portfolio via Bayesian multiple testing methodol…

PAC-Bayesian matrix completion with a spectral scaled Student prior offers efficient inference.

New RLHF framework handles general preference oracles without reward functions.

We study the computational tractability of PAC reinforcement learning with rich observations. We present new provably sample-efficient algorithms for environments with deterministic hidden state dynamics and stochastic rich observations. These methods operate in an oracle model of computation -- accessing policy and va…

Statsformer validates and adapts LLM-derived semantic priors for improved supervised learning.

New algorithm learns efficiently with a simple 'yes/no' oracle.

An algorithm learns from multiple models to match an oracle's risk.

Improved algorithm for contextual bandits with reduced regret.

New meta-reinforcement learning method improves performance in finite-horizon MDPs.

The paper analyzes sparse high-dimensional linear regression with random design and unknown error variance, providing adaptiveness and concentration rates.

New method improves posterior sampling for complex data models.

Oracle-efficient algorithm for offline RL with partial data coverage.

Adapts to misspecification in contextual bandits using offline regression.

Quantum algorithms for multi-armed bandits are explored with limited reward access.

New tool for parallel and private stochastic convex optimization reduces query complexity.

Estimates CATEs for structured treatments using a new decomposition method.

New method removes oracle and reduces memory usage for robust MDPs.

Approximate Message Passing (AMP) has been shown to be an excellent statistical approach to signal inference and compressed sensing problem. The AMP framework provides modularity in the choice of signal prior; here we propose a hierarchical form of the Gauss-Bernouilli prior which utilizes a Restricted Boltzmann Machin…

We study the Nonparametric Maximum Likelihood Estimator (NPMLE) for estimating Gaussian location mixture densities in $d$-dimensions from independent observations. Unlike usual likelihood-based methods for fitting mixtures, NPMLEs are based on convex optimization. We prove finite sample results on the Hellinger accurac…

Improved stochastic Halpern iteration for fixed-point approximation in normed spaces.

We study the problem of generating adversarial examples in a black-box setting in which only loss-oracle access to a model is available. We introduce a framework that conceptually unifies much of the existing work on black-box attacks, and we demonstrate that the current state-of-the-art methods are optimal in a natura…

Unified approach for quantum and classical learning from evaluation oracles.

New algorithms adapt to model misspecification in contextual bandits.

Paper addresses hybrid learning with constrained adversaries, achieving optimal performance.

New bound improves on weighted majority vote risk estimation.

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

LIBO optimizes repeated bandit tasks without prior knowledge or regret.

In this article, we investigate large sample properties of model selection procedures in a general Bayesian framework when a closed form expression of the marginal likelihood function is not available or a local asymptotic quadratic approximation of the log-likelihood function does not exist. Under appropriate identifi…

RS-NSGD improves SGD convergence for heavy-tailed noise.

New algorithms find optimal policies without knowing MDP span.

We present a new method for design problems wherein the goal is to maximize or specify the value of one or more properties of interest. For example, in protein design, one may wish to find the protein sequence that maximizes fluorescence. We assume access to one or more, potentially black box, stochastic "oracle" predi…

New sampling methods for constrained and composite distributions.

We study an online decision making problem where on each round a learner chooses a list of items based on some side information, receives a scalar feedback value for each individual item, and a reward that is linearly related to this feedback. These problems, known as contextual semibandits, arise in crowdsourcing, rec…

EB-PCA reduces noise in high-dimensional PCA by estimating a joint prior distribution.

We investigate multiarmed bandits with delayed feedback, where the delays need neither be identical nor bounded. We first prove that "delayed" Exp3 achieves the $O(\sqrt{(KT + D)\ln K} )$ regret bound conjectured by Cesa-Bianchi et al. [2019] in the case of variable, but bounded delays. Here, $K$ is the number of actio…

Develops NPMC method for noisy labels, improving multiclass classification accuracy.

The paper improves Bayesian precision matrix estimation for high-dimensional sparse data.

Oracle-efficient algorithms reduce combinatorial semi-bandit regret to logarithmic time.

New algorithms sample structured logconcave families with improved efficiency.

This paper is about variable selection, clustering and estimation in an unsupervised high-dimensional setting. Our approach is based on fitting constrained Gaussian mixture models, where we learn the number of clusters $K$ and the set of relevant variables $S$ using a generalized Bayesian posterior with a sparsity indu…

New method improves CATE model selection with optimal regret rates.

New RL method reduces sample complexity for large policy spaces.

New analysis shows Thompson Sampling can work with greedy approximations in combinatorial bandits.

Financial markets are complex environments that produce enormous amounts of noisy and non-stationary data. One fundamental problem is online portfolio selection, the goal of which is to exploit this data to sequentially select portfolios of assets to achieve positive investment outcomes while managing risks. Various al…

New algorithms sample convex bodies using Markov chains and restricted Gaussian oracles.