On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

count % of corpus

15 results for “OFUL”

Bandits with Mean Boundscs.LG

New algorithms exploit mean bounds to improve bandit problem performance.

problem Improving bandit problem performance with side information on arm means.

method Developed novel algorithms R-OFUL and GLUE exploiting mean bounds for tighter estimates and reduced exploration.

result Regret bounds for R-OFUL and GLUE are never worse than standard algorithms, demonstrating improved performance.

bandit problem mean bounds subgaussian estimates regret analysis algorithm development

2020-02-19abs ↗pdf ↗

Meta-learning with Stochastic Linear Banditsstat.ML

Meta-learning improves performance in stochastic linear bandits.

problem Selecting a learning algorithm that performs well across multiple bandit tasks.

method Regularized OFUL algorithm with a bias vector, estimating bias within the learning-to-learn setting.

result Meta-learning strategies improve performance when the number of tasks grows and task variance is small.

meta-learning stochastic linear bandits regression OFUL bias estimation

2020-05-18abs ↗pdf ↗

A General Theory of the Stochastic Linear Bandit and Its Applicationscs.LG

Paper introduces a new analysis framework for stochastic linear bandits.

problem Optimizing decision-making in online experiments with noisy rewards.

method Develops a general analysis framework and algorithms for stochastic linear bandits.

result Introduces new algorithms like SG that improve performance and provide new regret bounds.

stochastic linear bandits regret bounds optimism in expectation contextual MABs

2020-02-12abs ↗pdf ↗

Stochastic Online Instrumental Variable Regression: Regrets for Endogeneity and Bandit Feedbackcs.LG

New method tackles endogeneity in online learning with improved regret bounds.

problem Endogeneity in real data due to omitted variables, strategic behaviors, etc.

method O2SLS (Online Two-Stage Least Squares) for Instrumental Variable (IV) regression.

result O2SLS achieves identification and oracle regret bounds for stochastic online learning.

Instrumental Variable Endogeneity Stochastic Online Learning Oracle Regret Bandit Feedback

2023-02-18abs ↗pdf ↗

Efficient Multinomial Logistic Bandit via Frequent Directionscs.LG

A new algorithm reduces the time and space complexity for multinomial logistic bandits.

problem High-dimensional feedback in multinomial logistic bandits makes existing algorithms inefficient.

method Integrates frequent directions matrix sketching into OFUL-MLogB to reduce time and space complexity.

result Achieves a regret bound of

ilde{\mathcal{O}}(Δ_T(Kd\lnΔ_T+m)\sqrt{T})

multinomial logistic bandits Frequent Directions UCB-type algorithms regret bound online Newton updates

2026-06-10abs ↗pdf ↗

Efficient Linear Bandits through Matrix Sketchingcs.LG

We prove that two popular linear contextual bandit algorithms, OFUL and Thompson Sampling, can be made efficient using Frequent Directions, a deterministic online sketching technique. More precisely, we show that a sketch of size $m$ allows a $\mathcal{O}(md)$ update time for both algorithms, as opposed to $Ω(d^2)$ req…

2018-09-28abs ↗pdf ↗

Geometry-Aware Approaches for Balancing Performance and Theoretical Guarantees in Linear Banditscs.LG

New technique tracks geometric properties to correct base algorithms' poor performance.

problem Discrepancy between empirical performance and theoretical regret bounds in linear bandits.

method Data-driven technique that incorporates geometric information to formulate frequentist regret bound.

result Course-corrected algorithms achieve minimax optimal regret of

ilde{\mathcal{O}}(d\sqrt{T})

linear bandits regret bounds geometric properties Thompson sampling Greedy

2023-06-26abs ↗pdf ↗

Linear Bandits beyond Inner Product Spaces, the case of Bandit Optimal Transportstat.ML

New algorithm for linear bandits tackles Optimal Transport problems.

problem Optimal Transport problems not covered by traditional linear bandits.

method Embed actions into a Hilbertian subspace, penalize optimism, use least-squares estimation.

result Achieves same regret bounds as OFUL but interpolates between

ilde{\mathcal O}(\sqrt{T})

and

{\mathcal O}(T)

Linear Bandits Optimal Transport Kantorovich formulation Hilbertian Subspace Confidence Sets

2025-02-11abs ↗pdf ↗

Bandit algorithms to emulate human decision making using probabilistic distortionscs.LG

Motivated by models of human decision making proposed to explain commonly observed deviations from conventional expected value preferences, we formulate two stochastic multi-armed bandit problems with distorted probabilities on the reward distributions: the classic $K$ -armed bandit and the linearly parameterized bandit…

2016-11-30abs ↗pdf ↗

Federated Linear Bandits with Finite Adversarial Actionscs.LG

A federated learning algorithm tackles linear bandits with adversarial actions, achieving optimal regret bounds.

problem Federated linear bandits with finite adversarial action sets.

method FedSupLinUCB algorithm, extending SupLinUCB and OFUL principles.

result Achieves a total regret of

ilde{O}(\sqrt{d T})

, matching minimax lower bound and being order-optimal.

federated learning linear bandits adversarial actions supLinUCB OFUL

2023-11-02abs ↗pdf ↗

Bilinear Bandits with Low-rank Structurecs.LG

We introduce the bilinear bandit problem with low-rank structure in which an action takes the form of a pair of arms from two different entity types, and the reward is a bilinear function of the known feature vectors of the arms. The unknown in the problem is a $d_1$ by $d_2$ matrix $\mathbfΘ^*$ that defines the reward…

2019-01-08abs ↗pdf ↗

Linear Bandit algorithms using the Bootstrapstat.ML

This study presents two new algorithms for solving linear stochastic bandit problems. The proposed methods use an approach from non-parametric statistics called bootstrapping to create confidence bounds. This is achieved without making any assumptions about the distribution of noise in the underlying system. We present…

2016-05-04abs ↗pdf ↗

Concentrated Differential Privacy for Banditsstat.ML

Paper addresses DP in bandits, focusing on zCDP and providing private algorithms.

problem Privacy concerns in recommender systems using user-sensitive data.

method Formalizes and compares different DP adaptations to bandits, proposes private algorithms for various bandit settings.

result Private algorithms ensure negligible privacy costs compared to non-private regret.

Differential Privacy Bandits Concentrated Differential Privacy Recommender Systems Privacy

2023-09-01abs ↗pdf ↗

Stochastic Linear Bandits with Partially Observed Actionscs.LG

TOFU-POV tackles partially observed linear bandits, achieving sublinear regret with low-dimensional action vectors.

problem Stochastic linear bandits with partially observed actions in settings like recommendation and healthcare.

method TOFU-POV estimates latent action subspace, imputes missing actions, and runs OFUL in low-dimensional coordinates.

result TOFU-POV achieves

\sqrt{T}

regret scaling with intrinsic subspace dimension, improving upon natural baselines.

stochastic linear bandits partial observability low intrinsic dimension sublinear regret recommendation systems

2026-07-09abs ↗pdf ↗

Tackling Heavy-Tailed Rewards in Reinforcement Learning with Function Approximation: Minimax Optimal and Instance-Dependent Regret Boundscs.LG

New algorithm tackles heavy-tailed rewards in RL with instance-dependent regret bounds.

problem Efficient algorithms for RL with heavy-tailed rewards in large state-action spaces.

method Design of \textsc{Heavy-OFUL} for heavy-tailed linear bandits and \textsc{Heavy-LSVI-UCB} for RL with linear function approximation.

result First instance-dependent regret bounds for heavy-tailed rewards in RL with linear function approximation.

Reinforcement Learning Heavy-tailed Rewards Function Approximation Minimax Optimal Instance-Dependent Regret

2023-06-12abs ↗pdf ↗