Aug 2026· International Conference Computational Vision and Bio Inspired Computing· pp. 41-47· 0 citations· 30 references
Abstract
Recent advances in learning-augmented streaming show that access to an $\epsilon$-accurate oracle enables single-pass sublinear-space algorithms to surpass the classical 1/2 approximation barrier for MAX-CUT, achieving guarantees of $1 / 2+\Omega\left(\epsilon^{2}\right)$. However, existing formulations implicitly assume uniform oracle accuracy, benign stream orderings, and stationarity, limiting applicability to real-world graph streams exhibiting heavy-tailed sparsity, adversarial reordering, and temporal drift. This paper introduces a unified framework with three components: a Sparsity-Adaptive Hierarchical Oracle (SAHO), an Adversarially Robust Stream Processing (ARSP) mechanism, and a Temporal Community-Aware Oracle (TCAO). Theoretical analysis establishes that the framework generalizes uniformoracle learning-augmented streaming. Empirical evaluations on heavy-tailed communication networks and bounded-degree spatial graphs confirm consistent gains over uniform baselines while matching classical greedy guarantees when oracle information provides no marginal benefit.
This work proposes the first robust estimator achieving near-optimal dimension-free statistical rates in this setting, and extends the trimmed-mean framework underlying recent advances in robust mean and covariance estimation to arbitrary tensor order.
R. Oliveira, Zoraida F. Rico, Philip Thompson· 0 citations
Online learning is an attractive framework in many domains because it permits well-defined learning even when data are dependent or chosen adversarially. This generality, however, comes at a steep price, introducing significant statistical and computational barriers. Recently, smoothed online learning has emerged as a...
S. Voitovych, Adam Block, Alexander Rakhlin et al.· 0 citations
We study exact support recovery under $(\epsilon,\delta)$-differential privacy in sparse high-dimensional linear regression. We introduce Saturated Propose-test-release, a general mechanism that privately releases the output of a discrete selector with probability one once its stability certificate reaches a finite thr...
A penalized DRO formulation in which the adversary may choose any distribution but incurs a Wasserstein penalty for deviating from the empirical distribution is studied, showing that the adversary's problem can be reformulated as an optimization problem over transport maps that push empirical samples to adversarial one...
Alireza Abdollahpoorrostam, Ehsan Sharifian, Buse Sen et al.· 0 citations
This work studies online maximization of nonnegative, non-monotone DR-submodular functions over compact convex down-closed subsets of the $d$-dimensional unit cube and shows that the best known constructive offline approximation factor is also achievable online.
It is shown that a simple Follow-the-Perturbed-Leader algorithm with Gaussian perturbation for each observed context achieves the optimal $\widetilde O(\sqrt{T\log N})$ expected regret for a class of $N$ experts, while requiring one optimization-oracle call per round and no explicit enumeration of the class.
G. Buzaglo, Elad Hazan· 1 citation
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
