We are interested in a framework of online learning with kernels for low-dimensional, but large-scale and potentially adversarial datasets.

We give two new algorithms for this problem in the informed setting.

Neural Information Processing Systems http://nips.cc/

We present a computationally efficient algorithm for learning in this framework that simultaneously achieves near-optimal regret bounds in both stochastic and adversarial environments.

We present a computationally efficient algorithm for learning in this framework that simultaneously achieves near-optimal regret bounds in both stochastic and adversarial environments.

The standard metric to quantify an online learning algorithm's performance is the regret.

We formalize sequential decision-making with information acquisition as the probing-augmented user-centric selection (PUCS) framework, where a learner first probes a subset of arms to obtain side information on resources and rewards, and then assigns $K$ plays to $M$ arms. PUCS covers applications such as ridesharing, wireless scheduling, and content recommendation, in which both resources and payoffs are initially unknown and probing is costly. For the offline setting with known distributions, we present a greedy probing algorithm with a constant-factor approximation guarantee $ζ= (e-1)/(2e-1)$. For the online setting with unknown distributions, we introduce OLPA, a stochastic combinatorial bandit algorithm that achieves a regret bound $\mathcal{O}(\sqrt{T} + \ln^{2} T)$. We also prove a lower bound $Ω(\sqrt{T})$, showing that the upper bound is tight up to logarithmic factors. Experiments on real-world data demonstrate the effectiveness of our solutions.

The centralized collection of search interaction logs for training ranking models raises significant privacy concerns. Federated Online Learning to Rank (FOLTR) offers a privacy-preserving alternative by enabling collaborative model training without sharing raw user data. However, benchmarks in FOLTR are largely based on random partitioning of classical learning-to-rank datasets, simulated user clicks, and the assumption of synchronous client participation. This oversimplifies real-world dynamics and undermines the realism of experimental results. We present AOL4FOLTR, a large-scale web search dataset with 2.6 million queries from 10,000 users. Our dataset addresses key limitations of existing benchmarks by including user identifiers, real click data, and query timestamps, enabling realistic user partitioning, behavior modeling, and asynchronous federated learning scenarios.

We study the online learning with feedback graphs framework introduced by Man-nor and Shamir [24], in which the feedback received by the online learner is specified by a graphnull over the available actions.

We investigate the multiscale extension of the problem where the loss ranges of the experts are vastly different.