A robust variant of binary search achieves regret O(C + log T) for dynamic pricing with known corruption C and O(C + log² T) when unknown.
hub
Proceedings of the 19th international conference on World wide web , pages=
12 Pith papers cite this work. Polarity classification is still indexing.
12
Pith papers citing it
hub tools
citation-role summary
other 2
citation-polarity summary
roles
other 2polarities
unclear 2representative citing papers
Active sampling with allocation q_j proportional to p_j to the 2/3 achieves tight regret sqrt(n/T) times norm of p to the 2/3 for known context distribution p, with improvement up to Theta(k to the 1/4) over passive sampling.
Anchor-TS defines arm indices as the median of an online posterior sample, a hybrid posterior sample, and the online sample mean to correct distribution-shift bias and safely accelerate online learning with offline data.
Sub-network Laplace approximations always underestimate full-model predictive variance, and two new gradient-based and greedy selection rules provide theoretically grounded improvements.
A modular reduction from budget-constrained contextual bandits with adversarial contexts to unconstrained bandits via surrogate rewards, yielding improved guarantees and an efficient algorithm based on SquareCB.
Online learning algorithms for bidding in repeated second-price auctions achieve rate-optimal regret by modeling ad value as a causal treatment effect and exploiting second-price payment information.
NonZero introduces an interaction score and bandit-formalized proposal rule for local agent deviations in multi-agent MCTS, delivering a sublinear local-regret guarantee and improved sample efficiency on game benchmarks without full joint-action enumeration.
Adaptive Instruction Composition uses a neural contextual bandit with RL to adaptively combine crowdsourced texts, generating more effective and diverse LLM jailbreaks than random or prior adaptive methods on Harmbench.
A projection-based algorithm for COCO achieves O(log T) regret and O(log T) CCV for strongly convex losses and O(sqrt(T)) for convex losses by leveraging self-contracted curves.
Replaces determinant growth with generalized Rayleigh quotient for rare switching in private linear bandits to control worst-direction volume despite non-monotonic design matrices from noise.
Offline RL promises to extract high-utility policies from static datasets but faces fundamental challenges that current methods only partially address.
citing papers explorer
-
Toward Optimal Regret in Robust Pricing: Decoupling Corruption and Time
A robust variant of binary search achieves regret O(C + log T) for dynamic pricing with known corruption C and O(C + log² T) when unknown.
-
Active Context Selection Improves Simple Regret in Contextual Bandits
Active sampling with allocation q_j proportional to p_j to the 2/3 achieves tight regret sqrt(n/T) times norm of p to the 2/3 for known context distribution p, with improvement up to Theta(k to the 1/4) over passive sampling.
-
Sample-Mean Anchored Thompson Sampling for Offline-to-Online Learning with Distribution Shift
Anchor-TS defines arm indices as the median of an online posterior sample, a hybrid posterior sample, and the online sample mean to correct distribution-shift bias and safely accelerate online learning with offline data.
-
Optimality of Sub-network Laplace Approximations: New Results and Methods
Sub-network Laplace approximations always underestimate full-model predictive variance, and two new gradient-based and greedy selection rules provide theoretically grounded improvements.
-
Constrained Contextual Bandits with Adversarial Contexts
A modular reduction from budget-constrained contextual bandits with adversarial contexts to unconstrained bandits via surrogate rewards, yielding improved guarantees and an efficient algorithm based on SquareCB.
-
The (Marginal) Value of a Search Ad: An Online Causal Framework for Repeated Second-price Auctions
Online learning algorithms for bidding in repeated second-price auctions achieve rate-optimal regret by modeling ad value as a causal treatment effect and exploiting second-price payment information.
-
NonZero: Interaction-Guided Exploration for Multi-Agent Monte Carlo Tree Search
NonZero introduces an interaction score and bandit-formalized proposal rule for local agent deviations in multi-agent MCTS, delivering a sublinear local-regret guarantee and improved sample efficiency on game benchmarks without full joint-action enumeration.
-
Adaptive Instruction Composition for Automated LLM Red-Teaming
Adaptive Instruction Composition uses a neural contextual bandit with RL to adaptively combine crowdsourced texts, generating more effective and diverse LLM jailbreaks than random or prior adaptive methods on Harmbench.
-
Improved Guarantees for Constrained Online Convex Optimization via Self-Contraction
A projection-based algorithm for COCO achieves O(log T) regret and O(log T) CCV for strongly convex losses and O(sqrt(T)) for convex losses by leveraging self-contracted curves.
-
When Determinants Are Not Enough: Private Rare Switching
Replaces determinant growth with generalized Rayleigh quotient for rare switching in private linear bandits to control worst-direction volume despite non-monotonic design matrices from noise.
-
Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems
Offline RL promises to extract high-utility policies from static datasets but faces fundamental challenges that current methods only partially address.
- Online Learning-to-Defer with Varying Experts