520 Hits in 1.3 sec

Privacy-Preserving Multi-Party Contextual Bandits [article]

Awni Hannun, Brian Knott, Shubho Sengupta, Laurens van der Maaten
2020 arXiv   pre-print
This paper develops a privacy-preserving multi-party contextual bandit for this learning setting by combining secure multi-party computation with a differentially private mechanism based on epsilon-greedy  ...  Contextual bandits are commonly used to solve recommendation or ranking problems.  ...  Privacy-Preserving Multi-Party Contextual Bandits  ... 
arXiv:1910.05299v3 fatcat:4yt2qxezifgo3ofqylodcpa6cy

Locally Differentially Private (Contextual) Bandits Learning [article]

Kai Zheng, Tianle Cai, Weiran Huang, Zhenguo Li, Liwei Wang
2021 arXiv   pre-print
Note that given the existing Ω(T) lower bound for DP contextual linear bandits (Shariff Sheffe, 2018), our result shows a fundamental difference between LDP and DP contextual bandits learning.  ...  Based on our frameworks, we can improve previous best results for private bandits learning with one-point feedback, such as private Bandits Convex Optimization, and obtain the first result for Bandits  ...  Note the non-linearity of g makes things much more complicated either from the view of bandits learning or privacy preservation.  ... 
arXiv:2006.00701v4 fatcat:quapec7ss5fl7hlzzurpbvin3i

Privacy-Preserving Bandits [article]

Mohammad Malekzadeh, Dimitrios Athanasakis, Hamed Haddadi, Benjamin Livshits
2020 arXiv   pre-print
Contextual bandit algorithms (CBAs) often rely on personal data to provide recommendations.  ...  This paper proposes a technique we call Privacy-Preserving Bandits (P2B); a system that updates local agents by collecting feedback from other local agents in a differentially-private manner.  ...  CONCLUSIONS This paper presents P2B, a privacy-preserving approach for machine learning with contextual bandits.  ... 
arXiv:1909.04421v4 fatcat:ynffzmb3czc33dxlkncocevyja

Dynamic Privacy Pricing For Timely Rewards

Shatabdi Nandi
2018 International Journal for Research in Applied Science and Engineering Technology  
Conclusion-It is useful to protect individual's privacy and to set the proper pay off.  ...  Setting a price for individual's privacy is one form to conquer these threats is a measure though it is a tough issue.  ...  Contextual Bandit Approach Instead of estimating the cumulative distribution, here we view the time-variant characteristic of the bandit problem from a different perspective. G.  ... 
doi:10.22214/ijraset.2018.2134 fatcat:npvtposimnevti2qffjso226pa

Locally Differentially Private Reinforcement Learning for Linear Mixture Markov Decision Processes [article]

Chonghua Liao and Jiafan He and Quanquan Gu
2021 arXiv   pre-print
To protect the users' privacy, privacy-preserving RL algorithms are in demand. In this paper, we study RL with linear function approximation and local differential privacy (LDP) guarantees.  ...  To the best of our knowledge, this is the first provable privacy-preserving RL algorithm with linear function approximation.  ...  In the LDP setting, the privacy-preserving mechanism M generates the privatized version of the context x t , denoted by r x t " Mpx t q, to the contextual linear bandit algorithm.  ... 
arXiv:2110.10133v1 fatcat:gvvkdlzxr5eylcawjdp5xrmgcy

Mitigating Bias in Adaptive Data Gathering via Differential Privacy [article]

Seth Neel, Aaron Roth
2018 arXiv   pre-print
hypothesis tests on complex data gathered via contextual bandit algorithms leads to false discovery.  ...  Moreover, there exist differentially private bandit algorithms with near optimal regret bounds: we apply existing theorems in the simple stochastic case, and give a new analysis for linear contextual bandits  ...  Contextual Bandit Problems In the contextual bandit problem, decisions are endowed with observable features.  ... 
arXiv:1806.02329v1 fatcat:jj7jifrysjd35l3f52umsnzj54

Differentially Private Contextual Linear Bandits [article]

Roshan Shariff, Or Sheffet
2018 arXiv   pre-print
Our goal is to devise private learners for the contextual linear bandit problem. We first show that using the standard definition of differential privacy results in linear regret.  ...  We study the contextual linear bandit problem, a version of the standard stochastic multi-armed bandit (MAB) problem where a learner sequentially selects actions to maximize a reward which depends also  ...  [1] gives an instance dependent bound for linear bandits, which we convert to the contextual setting. Differential Privacy. Differential privacy, first introduced by Dwork et al.  ... 
arXiv:1810.00068v1 fatcat:rdvsvyj54ratlkstlxqfebupt4

Privacy-Preserving Dynamic Personalized Pricing with Demand Learning [article]

Xi Chen and David Simchi-Levi and Yining Wang
2021 arXiv   pre-print
Using the fundamental framework of differential privacy from computer science, we develop a privacy-preserving dynamic pricing policy, which tries to maximize the retailer revenue while avoiding information  ...  Our policy achieves both the privacy guarantee and the performance guarantee in terms of regret.  ...  In fact, this is still an open problem for generalized linear contextual bandit under the DP guarantee.  ... 
arXiv:2009.12920v2 fatcat:ibwk2m4ptrdxdhnptlal262loe

Federated Bandit: A Gossiping Approach [article]

Zhaowei Zhu, Jingxuan Zhu, Ji Liu, Yang Liu
2020 arXiv   pre-print
We then propose Fed_UCB, a differentially private version of Gossip_UCB, in which the agents preserve ϵ-differential privacy of their local data while achieving O(max{poly(N,M)/ϵlog^2.5 T, poly(N,M) (log_λ  ...  In this paper, we study Federated Bandit, a decentralized Multi-Armed Bandit problem with a set of N agents, who can only communicate their local data with neighbors described by a connected graph G.  ...  Future work may include extending this framework to contextual bandits [46] with local features or bandits with continuous arms [53] .  ... 
arXiv:2010.12763v1 fatcat:skq7homjy5cs5phj3ikbcda2r4

Online learning with Corrupted context: Corrupted Contextual Bandits [article]

Djallel Bouneffouf
2020 arXiv   pre-print
In order to address the corrupted-context setting,we propose to combine the standard contextual bandit approach with a classical multi-armed bandit mechanism.  ...  We consider a novel variant of the contextual bandit problem (i.e., the multi-armed bandit with side-information, or context, available to a decision-maker) where the context used at each decision may  ...  In this framework, motivated by privacy preserving in online recommender systems, the goal is to maximize the sum of the (unobserved) rewards, based on the observation of transformation of these rewards  ... 
arXiv:2006.15194v1 fatcat:aipupxx235gybdgytq2cnfbcue

Cascading Bandit under Differential Privacy [article]

Kun Wang, Jing Dong, Baoxiang Wang, Shuai Li, Shuo Shao
2021 arXiv   pre-print
This paper studies differential privacy (DP) and local differential privacy (LDP) in cascading bandits.  ...  Our results extend to combinatorial semi-bandit. We show respective lower bounds for DP and LDP cascading bandits. Extensive experiments corroborate our theoretic findings.  ...  Conservative contextual combinatorial cascading bandit. arXiv preprint arXiv:2104.08615, 2021.  ... 
arXiv:2105.11126v2 fatcat:mwmdsqfelrhh7ni3onk5h4vzxe

Multi-Armed Bandits with Local Differential Privacy [article]

Wenbo Ren, Xingyu Zhou, Jia Liu, Ness B. Shroff
2020 arXiv   pre-print
This paper investigates the problem of regret minimization for multi-armed bandit (MAB) problems with local differential privacy (LDP) guarantee.  ...  To handle this dilemma, we adopt differential privacy and study the regret upper and lower bounds for MAB algorithms with a given LDP guarantee.  ...  In [25], the authors studied privacy-preserving adversarial bandits.  ... 
arXiv:2007.03121v1 fatcat:eehdco6p4ndpffatetx47o4nzi

Differential Privacy for Multi-armed Bandits: What Is It and What Is Its Cost? [article]

Debabrota Basu, Christos Dimitrakakis, Aristide Tossou
2020 arXiv   pre-print
Based on differential privacy (DP) framework, we introduce and unify privacy definitions for the multi-armed bandit algorithms.  ...  We derive and contrast lower bounds on the regret of bandit algorithms satisfying these definitions. We leverage a unified proving technique to achieve all the lower bounds.  ...  Shariff and Sheffet (2018) proves a finite-time problem-dependent lower bound for contextual bandits.  ... 
arXiv:1905.12298v2 fatcat:eiv2s4b7yrgajjr2vl3m6z2udi

Privacy-Aware Online Task Offloading for Mobile-Edge Computing [chapter]

Ting Li, Haitao Liu, Jie Liang, Hangsheng Zhang, Liru Geng, Yinlong Liu
2020 Lecture Notes in Computer Science  
Firstly, we formulate the joint optimization problem of task offloading and privacy preservation as a semiparametric contextual multi-armed bandit (MAB) problem, which has a relaxed reward model.  ...  In order to tackle these challenges, a privacy-preserving and device-managed task offloading scheme is proposed in this paper for MEC.  ...  In the cloud layer, authors proposed a privacy-preserving and contextual online learning algorithm to manage the participants' reputation.  ... 
doi:10.1007/978-3-030-59016-1_21 fatcat:wmjd24uhurgrdl4xq4udbv6zla

Federated Linear Contextual Bandits [article]

Ruiquan Huang, Weiqiang Wu, Jing Yang, Cong Shen
2021 arXiv   pre-print
This paper presents a novel federated linear contextual bandits model, where individual clients face different K-armed stochastic bandits coupled through common global parameters.  ...  Li et al. (2020) and Zhu et al. (2021) focus on differential privacy based local data privacy protection in federated bandits.  ...  Li et al. (2020) and Zhu et al. (2021) focus on differential privacy based local data privacy protection in federated bandits.  ... 
arXiv:2110.14177v1 fatcat:y322gkg7cvfcrnqckr5lfrdljm
« Previous Showing results 1 — 15 out of 520 results