Tag: bound
-
Online Learning of Neural Networks
Online Learning of Neural Networks arXiv:2505.09167v1 Announce Type: new Abstract: We study online learning of feedforward neural networks with the sign activation function that implement functions from the unit ball in $mathbb{R}^d$ to a finite label set ${1, ldots, Y}$. First, we characterize a margin condition that is sufficient and in some cases necessary for…
-
Lower Bounds for Greedy Teaching Set Constructions
Lower Bounds for Greedy Teaching Set Constructions arXiv:2505.03223v1 Announce Type: new Abstract: A fundamental open problem in learning theory is to characterize the best-case teaching dimension $operatorname{TS}_{min}$ of a concept class $mathcal{C}$ with finite VC dimension $d$. Resolving this problem will, in particular, settle the conjectured upper bound on Recursive Teaching Dimension posed by [Simon…
-
Empirical Bound Information-Directed Sampling for Norm-Agnostic Bandits
Empirical Bound Information-Directed Sampling for Norm-Agnostic Bandits arXiv:2503.05098v1 Announce Type: new Abstract: Information-directed sampling (IDS) is a powerful framework for solving bandit problems which has shown strong results in both Bayesian and frequentist settings. However, frequentist IDS, like many other bandit algorithms, requires that one have prior knowledge of a (relatively) tight upper bound on…
-
Improved Online Confidence Bounds for Multinomial Logistic Bandits
Improved Online Confidence Bounds for Multinomial Logistic Bandits arXiv:2502.10020v1 Announce Type: new Abstract: In this paper, we propose an improved online confidence bound for multinomial logistic (MNL) models and apply this result to MNL bandits, achieving variance-dependent optimal regret. Recently, Lee & Oh (2024) established an online confidence bound for MNL models and achieved nearly…