Distributed Decision-Making Over Adaptive Networks

Abstract

In distributed processing, agents generally collect data generated by the same underlying unknown model (represented by a vector of parameters) and then solve an estimation or inference task cooperatively.In this paper, we consider the situation in which the data observed by the agents may have risen from two different models. Agents do not know beforehand which model accounts for their data and the data of their neighbors. The objective for the network is for all agents to reach agreement on which model to track and to estimate this model cooperatively. In these situations, where agents are subject to data from unknown different sources, conventional distributed estimation strategies would lead to biased estimates relative to any of the underlying models. We first show how to modify existing strategies to guarantee unbiasedness. We then develop a classification scheme for the agents to identify the models that generated the data, and propose a procedure by which the entire network can be made to converge towards the same model through a collaborative decision-making process. The resulting algorithm is applied to model fish foraging behavior in the presence of two food sources.

Distributed Clustering and Learning Over Networks

“…Although cluster information is in general not available beforehand, groups within each cluster are available according to Assumption 1. Using this prior information, agents can instead focus on solving the following problem based on partitioning by groups rather than by clusters: (19) with one parameter vector for each group . In the extreme case when prior clustering information is totally absent, groups will collapse into singletons and problem (19) will reduce to the individual non-cooperative case with each agent running its own stochastic-gradient algorithm to minimize its cost function.…”

“…This early investigation dealt only with the case of two separate clusters in the network with each cluster interested in one parameter vector. One useful application of this formulation in the context of biological networks was considered in [19], where each agent was assumed to collect data arising from one of two models (e.g., the location of two separate food sources). The agents did not know which model generated their observations and, yet, they needed to reach agreement about which model to follow (i.e., which food source to move towards).…”

Adaptive regularized diffusion adaptation over multitask networks

“…We therefore focus on the implementation of diffusion strategies in this article. In particular, we examine networks where different clusters of agents may be interested in different objectives [10][11][12][13][14][15]. In this case, it is important to develop algorithms that enable the agents to continuously learn which of their neighbors belong to the same cluster and which ones are from different clusters.…”

Robust and computationally efficient diffusion-based classification in distributed networks

“…This motivates looking for in-network classification algorithms where a minimum amount of information is exchanged among single-hop neighbors. Although several methods have been proposed in the last years that deal with distributed data clustering and classification [2,3,4,5,6,7,8,9,10,11], most of them still assume the presence of a fusion center [5,7], are hardly real-time capable [3] or need a set of prelabelled training data for training beforehand [6,11]. Several distributed adaptive strategies, such as incremental, consensus, and diffusion algorithms have been developed in the last few years.…”