Generalized Linear Bandits with Local Differential Privacy
6. prosince 2021
Řečníci
Zhipeng Liang
Řečník · 0 sledujících
O prezentaci
Contextual bandit algorithms are useful in personalized online decision-making. However, many applications such as personalized medicine and online advertising require the utilization of individual-specific information for effective learning, while user's data should remain private from the server due to privacy concerns. This motivates the introduction of local differential privacy (LDP), a stringent notion in privacy, to contextual bandits. In this paper, we design LDP algorithms for stochastic generalized linear bandits to achieve the same regret bound as in non-privacy settings. Our main idea is to develop a stochastic gradient-based estimator and update mechanism to ensure LDP. We then exploit the flexibility of stochastic gradient descent (SGD), whose theoretical guarantee for bandit problems is rarely explored, in dealing with generalized linear bandits. We also develop an estimator and update mechanism based on Ordinary Least Square (OLS) for linear bandits. Finally, we conduct experiments with both simulation and real-world datasets to demonstrate the consistently superb performance of our algorithms under LDP constraints with reasonably small parameters (ε, δ) to ensure strong privacy protection.Contextual bandit algorithms are useful in personalized online decision-making. However, many applications such as personalized medicine and online advertising require the utilization of individual-specific information for effective learning, while user's data should remain private from the server due to privacy concerns. This motivates the introduction of local differential privacy (LDP), a stringent notion in privacy, to contextual bandits. In this paper, we design LDP algorithms for stochasti…
Organizátor
NeurIPS 2021
Účet · 1,9k sledujících
O organizátorovi (NeurIPS 2021)
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Baví vás formát? Nechte SlidesLive zachytit svou akci!
Profesionální natáčení a streamování po celém světě.
Sdílení
Doporučená videa
Prezentace na podobné téma, kategorii nebo přednášejícího
OctField: Hierarchical Implicit Functions for 3D Modeling
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Shape your Space: A Gaussian Mixture Regularization Approach to Deterministic Autoencoders
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Pl@ntNet-300K: a plant image dataset with high label ambiguity and a long-tailed distribution
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Fair Scheduling for Time-dependent Resources
Zhlédnout později
Bo Li, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
SSMF: Shifting Seasonal Matrix Factorization
Zhlédnout později
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %
Turning high-throughput structural biology into predictive drug design
Zhlédnout později
Kadi Saar, …
Pro uložení prezentace do věčného trezoru hlasovalo 0 diváků, což je 0.0 %