IEEE TRANSACTIONS ON INFORMATION THEORY , VOL . ? , NO . ? , ? 200 ? 1 Improved Risk Tail Bounds for On-Line Algorithms

By Nicolò Cesa-Bianchi, Claudio Gentile. 2007

72 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

Citations in the corpus

Adaptive Gradient-Based Meta-Learning Methods (2019, neutral): “[13] Nicoló Cesa-Bianchi and Claudio Gentile.”
Introduction to Online Convex Optimization (2016, background): “Tighter and more general bounds were explored in (27; 28; 109).”
A Survey on Metric Learning for Feature Vectors and Structured Data (2013, background): “ion for online methods Wang et al. (2012d, 2013b) deal with the online learning setting. They show that existing proof techniques to convert regret bounds into generalization bounds (see for instance Cesa-Bianchi and Gentile, 2008) only hold for univariate loss functions, but derive an alternative f”
Online Machine Learning in Big Data Streams (2018, methodology): “ for the same instance. Batch gradient descent can be emulated by an online algorithm: We go through training examples one by one in an online fashion and repeat multiple times over the training data [45]. Based on the general idea of perceptron learning, several online linear models have been propo”
Online Learning: A Comprehensive Survey (2018, methodology): “Most of these algorithms explore the k classes uniformly with probability γ, while (Crammer and Gentile, 2011) sample the classes based on the Upper Confidence Bound.”
On the Generalization Ability of Online Strongly Convex Programming Algorithms (2008, methodology): “This paper continues the line of research initiated by several r searchers [Littlestone, 1989, CesaBianchi et al., 2004, Zhang, 2005, Cesa-Bianchi and Gentile, 2008] which looks at how to convert online algorithms into batch algorithms with provable guarantees.”
Provable Guarantees for Gradient-Based Meta-Learning (2019, methodology): “The result follows by nesting two different single-task online-to-batch conversions (Cesa-Bianchi et al., 2004; Cesa-Bianchi & Gentile, 2005) via Jensen’s inequality.”
Recursive Aggregation of Estimators by the Mirror Descent Algorithm with Averaging (2005, unknown): “However, these results are formulated in a deterministic setting and they do not extend straightforwardly to the standard stochastic analysis with a mean risk criterion (see [17, 8, 9] for insights on the connections between the two types of results).”
Online Clustering of Bandits (2014, unknown): “We apply Claim 1, so that λ min ( E 1 [ X 1 | c 1 ]) ≥ λ/ 4 , and proceed as in, e.g., (Cesa-Bianchi & Gentile, 2008).”