BAFFLE : Blockchain Based Aggregator Free Federated Learning
作者： Paritosh Ramanan , Kiyoshi Nakayama
DOI:
关键词:
摘要: A key aspect of Federated Learning (FL) is the requirement a centralized aggregator to maintain and update global model. However, in many cases orchestrating might be infeasible due numerous operational constraints. In this paper, we introduce BAFFLE, an free, blockchain driven, FL environment that inherently decentralized. BAFFLE leverages Smart Contracts (SC) coordinate round delineation, model aggregation tasks FL. boosts computational performance by decomposing parameter space into distinct chunks followed score bid strategy. order characterize conduct experiments on private Ethereum network use driven methods as our benchmark. We show significantly reduces gas costs for compared direct adaptation based method. Our results also achieves high scalability efficiency while delivering similar accuracy benchmark methods.
arxiv.org参考文章(25)
Martin Riedmiller, Neural Fitted Q Iteration – First Experiences with a Data Efficient Neural Reinforcement Learning Method Machine Learning: ECML 2005. pp. 317- 328 ,(2005) , 10.1007/11564096_32
Ali Khajeh-Hosseini, Ilango Sriram, Ian Sommerville, Research Challenges for Enterprise Cloud Computing arXiv: Distributed, Parallel, and Cluster Computing. ,(2010)
Edgar Gabriel, Graham E. Fagg, George Bosilca, Thara Angskun, Jack J. Dongarra, Jeffrey M. Squyres, Vishal Sahay, Prabhanjan Kambadur, Brian Barrett, Andrew Lumsdaine, Ralph H. Castain, David J. Daniel, Richard L. Graham, Timothy S. Woodall, Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation Lecture Notes in Computer Science. pp. 97- 104 ,(2004) , 10.1007/978-3-540-30218-6_19
Reza Shokri, Vitaly Shmatikov, Privacy-Preserving Deep Learning computer and communications security. pp. 1310- 1321 ,(2015) , 10.1145/2810103.2813687
Jakub Konečný, Daniel Ramage, H. Brendan McMahan, Federated Optimization: Distributed Optimization Beyond the Datacenter arXiv: Learning. ,(2015)
Konstantinos Christidis, Michael Devetsikiotis, Blockchains and Smart Contracts for the Internet of Things IEEE Access. ,vol. 4, pp. 2292- 2303 ,(2016) , 10.1109/ACCESS.2016.2566339
Jakub Konečný, Ananda Theertha Suresh, Dave Bacon, Felix X. Yu, Peter Richtarik, H. Brendan McMahan, Federated Learning: Strategies for Improving Communication Efficiency arXiv: Learning. ,(2016)
Miyoung Han, Pierre Senellart, Stéphane Bressan, Huayu Wu, Routing an Autonomous Taxi with Reinforcement Learning conference on information and knowledge management. pp. 2421- 2424 ,(2016) , 10.1145/2983323.2983379
Blaise Aguera y Arcas, Daniel Ramage, H. Brendan McMahan, Seth Hampson, Eider Moore, Communication-Efficient Learning of Deep Networks from Decentralized Data international conference on artificial intelligence and statistics. pp. 1273- 1282 ,(2017)
Marie Douriez, Harish Doraiswamy, Juliana Freire, Claudio T. Silva, Anonymizing NYC Taxi Data: Does It Matter? 2016 IEEE International Conference on Data Science and Advanced Analytics (DSAA). pp. 140- 148 ,(2016) , 10.1109/DSAA.2016.21