{"@context":"http://iiif.io/api/presentation/2/context.json","@id":"https://repo.library.stonybrook.edu/cantaloupe/iiif/2/manifest.json","@type":"sc:Manifest","label":"Distributed Bayesian Learning in Multi-agent Systems","metadata":[{"label":"dc.description.sponsorship","value":"This work is sponsored by the Stony Brook University Graduate School in compliance with the requirements for completion of degree."},{"label":"dc.format","value":"Monograph"},{"label":"dc.format.medium","value":"Electronic Resource"},{"label":"dc.identifier.uri","value":"http://hdl.handle.net/11401/77831"},{"label":"dc.language.iso","value":"en_US"},{"label":"dc.publisher","value":"The Graduate School, Stony Brook University: Stony Brook, NY."},{"label":"dcterms.abstract","value":"In the literature on multi-agent systems, the problems of how agents learn from messages of other agents and how based on all of their available information they make decisions have been widely studied. In this dissertation, we address three problems of Bayesian learning where through the Bayesian methodology, each agent makes inference on the state of nature from the messages of its neighboring agents. First, we study a system where the agents make decisions sequentially about the true state of nature. Each agent observes a signal produced according to one of two hypotheses. The agents also know the decisions of all the previous agents in the network. We consider the randomness in the agents decisions by introducing a random decision making policy. By analyzing the expected value of the agents beliefs, we prove that information cascade can be avoided and that asymptotic learning occurs. Second, we study the problem of distributed hypothesis testing in cooperative networks of agents over a given undirected graph. Each agent in the network has its private signal generated by one of two hypotheses. In each time slot, two agents are randomly selected to exchange their decisions. We propose a set of gossip-type methods for which two communicating agents reach the optimal local consensus with probability one by a few exchanges of binary actions at every time slot. We show that in a finite number of time slots, all the agents in the network will reach the optimal global consensus. Third, we study the problem of distributed Bayesian estimation. The agents observe data modeled by a general linear model and with covariance matrices of observation noise unknown to the agents. They try to reach consensus on the belief on the unknown linear parameters based on their private signals and information provided by their neighbors. We propose an information fusion and diffusion method for cooperative distributed estimation. We prove that with the proposed method, the Kullback-Leibler divergence between the beliefs of the agents and a fictitious fusion center converges to zero."},{"label":"dcterms.available","value":"2017-09-26T17:15:34Z"},{"label":"dcterms.contributor","value":"Bugallo, Monica"},{"label":"dcterms.creator","value":"Wang, Yunlong"},{"label":"dcterms.dateAccepted","value":"2017-09-26T17:15:34Z"},{"label":"dcterms.dateSubmitted","value":"2017-09-26T17:15:34Z"},{"label":"dcterms.description","value":"Department of Electrical Engineering."},{"label":"dcterms.extent","value":"183 pg."},{"label":"dcterms.format","value":"Application/PDF"},{"label":"dcterms.identifier","value":"Wang_grad.sunysb_0771E_12384.pdf"},{"label":"dcterms.issued","value":"2015-05-01"},{"label":"dcterms.language","value":"en_US"},{"label":"dcterms.provenance","value":"Submitted by Jason Torre (fjason.torre@stonybrook.edu) on 2017-09-26T17:15:34Z\nNo. of bitstreams: 1\nWang_grad.sunysb_0771E_12384.pdf: 1246531 bytes, checksum: 92cc6b2633a58588e34cdb724e679ae0 (MD5)"},{"label":"dcterms.publisher","value":"The Graduate School, Stony Brook University: Stony Brook, NY."},{"label":"dcterms.subject","value":"Asymptotic learning, Bayesian learning, Distributed estimation, Distributed hypothesis testing, Social learning"},{"label":"dcterms.title","value":"Distributed Bayesian Learning in Multi-agent Systems"},{"label":"dcterms.type","value":"Dissertation"},{"label":"dc.type","value":"Dissertation"}],"description":"This manifest was generated dynamically","viewingDirection":"left-to-right","sequences":[{"@type":"sc:Sequence","canvases":[{"@id":"https://repo.library.stonybrook.edu/cantaloupe/iiif/2/canvas/page-1.json","@type":"sc:Canvas","label":"Page 1","height":1650,"width":1275,"images":[{"@type":"oa:Annotation","motivation":"sc:painting","resource":{"@id":"https://repo.library.stonybrook.edu/cantaloupe/iiif/2/75%2F38%2F88%2F75388810341089805278314056395377553722/full/full/0/default.jpg","@type":"dctypes:Image","format":"image/jpeg","height":1650,"width":1275,"service":{"@context":"http://iiif.io/api/image/2/context.json","@id":"https://repo.library.stonybrook.edu/cantaloupe/iiif/2/75%2F38%2F88%2F75388810341089805278314056395377553722","profile":"http://iiif.io/api/image/2/level2.json"}},"on":"https://repo.library.stonybrook.edu/cantaloupe/iiif/2/canvas/page-1.json"}]}]}]}