A Distributed Privacy-Preserving Learning Dynamics in General Social Networks

In this paper, we study a distributed privacy-preserving learning problem in general social networks. Specifically, we consider a very general problem setting where the agents in a given multi-hop social network are required to make sequential decisions to choose among a set of options featured by unknown stochastic quality signals. Each agent is allowed to interact with its peers through multi-hop communications but with its privacy preserved. To serve the above goals, we propose a four-staged distributed social learning algorithm. In a nutshell, our algorithm proceeds iteratively, and in every round, each agent i) randomly perturbs its adoption for privacy-preserving purpose, ii) disseminates the perturbed adoption over the social network in a nearly uniform manner through random walking, iii) selects an option by referring to its peers' perturbed latest adoptions, and iv) decides whether or not to adopt the selected option according to its latest quality signal. By our solid theoretical analysis, we provide answers to two fundamental algorithmic questions about the performance of our four-staged algorithm: on one hand, we illustrate the convergence of our algorithm when there are a sufficient number of agents in the social network, each of which are with incomplete and perturbed knowledge as input; on the other hand, we reveal the quantitative trade-off between the privacy loss and the communication overhead towards the convergence. We also perform extensive simulations to validate our theoretical analysis and to verify the efficacy of our algorithm.

[1]  Ahmed Ahmim,et al.  Privacy-Preserving Schemes for Ad Hoc Social Networks: A Survey , 2016, IEEE Communications Surveys & Tutorials.

[2]  Nisheeth K. Vishnoi,et al.  A Distributed Learning Dynamics in Social Groups , 2017, PODC.

[3]  Junshan Zhang,et al.  Data Collection from Privacy-Aware Users in the Presence of Social Learning , 2019, 2019 57th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[4]  Sung-Wook Kim,et al.  Learning New Words from Keystroke Data with Local Differential Privacy , 2020, IEEE Transactions on Knowledge and Data Engineering.

[5]  Thomas M. Cover,et al.  The two-armed-bandit problem with time-invariant finite memory , 1970, IEEE Trans. Inf. Theory.

[6]  Masatoshi Yoshikawa,et al.  Quantifying Differential Privacy in Continuous Data Release Under Temporal Correlations , 2017, IEEE Transactions on Knowledge and Data Engineering.

[7]  Jiguo Yu,et al.  Fast Fault-Tolerant Sampling via Random Walk in Dynamic Networks , 2019, 2019 IEEE 39th International Conference on Distributed Computing Systems (ICDCS).

[8]  Yanjiao Chen,et al.  Privacy-Preserving Collaborative Model Learning: The Case of Word Vector Training , 2018, IEEE Transactions on Knowledge and Data Engineering.

[9]  Sarvar Patel,et al.  Practical Secure Aggregation for Privacy-Preserving Machine Learning , 2017, IACR Cryptol. ePrint Arch..

[10]  V. Climenhaga.  Markov chains and mixing times , 2013 .

[11]  Vitaly Shmatikov,et al.  Privacy-preserving deep learning , 2015, 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[12]  Alex Pentland,et al.  Human collective intelligence as distributed Bayesian inference , 2016, ArXiv.

[13]  Jun Wang,et al.  Portfolio Choices with Orthogonal Bandit Learning , 2015, IJCAI.

[14]  T. Seeley,et al.  Group decision making in swarms of honey bees , 1999, Behavioral Ecology and Sociobiology.

[15]  Kuang Xu,et al.  Reinforcement with Fading Memories , 2018, SIGMETRICS.

[16]  Mark Lubell,et al.  Beyond existence and aiming outside the laboratory: estimating frequency-dependent and pay-off-biased social learning strategies , 2008, Philosophical Transactions of the Royal Society B: Biological Sciences.

[17]  A. Cabrales.  Stochastic replicator dynamics , 2000 .

[18]  H. Vincent Poor,et al.  Federated Learning With Differential Privacy: Algorithms and Performance Analysis , 2019, IEEE Transactions on Information Forensics and Security.

[19]  Mauro Birattari,et al.  Multi-armed Bandit Formulation of the Task Partitioning Problem in Swarm Robotics , 2012, ANTS.

[20]  Rui Zhang,et al.  A Hybrid Approach to Privacy-Preserving Federated Learning , 2018, Informatik Spektrum.

[21]  Shibo He,et al.  Socially Privacy-Preserving Data Collection for Crowdsensing , 2020, IEEE Transactions on Vehicular Technology.

[22]  Joseph Henrich,et al.  Cultural transmission and the diffusion of innovations : adoption dynamics indicate that biased cultural transmission is the predominate force in behavioral change and much of sociocultural evolution , 2001 .

[23]  Nicole Immorlica,et al.  Diversity and Exploration in Social Learning , 2019, WWW.

[24]  Aaron Roth,et al.  Privately releasing conjunctions and the statistical query barrier , 2010, STOC '11.

[25]  D. Yorke,et al.  Word-of-Mouth Communications , 2015 .

[26]  Gábor Danner,et al.  Fully Distributed Privacy Preserving Mini-batch Gradient Descent Learning , 2015, DAIS.

[27]  Alessandro Panconesi,et al.  Concentration of Measure for the Analysis of Randomized Algorithms , 2009 .

[28]  Ming Li,et al.  PCKV: Locally Differentially Private Correlated Key-Value Data Collection with Optimized Utility , 2019, USENIX Security Symposium.

[29]  Glenn Ellison,et al.  Word-of-Mouth Communication and Social Learning , 1995 .

[30]  Payman Mohassel,et al.  SecureML: A System for Scalable Privacy-Preserving Machine Learning , 2017, 2017 IEEE Symposium on Security and Privacy (SP).

[31]  Elizabeth L. Wilmer,et al.  Markov Chains and Mixing Times , 2008 .

[32]  C. Lumsden.  Culture and the Evolutionary Process, Robert Boyd, Peter J. Richerson. University of Chicago Press, Chicago & London (1985), viii, +301. Price $29.95 , 1986 .