A General Framework for Distributed Inference with Uncertain Models
This work provides a generalized framework for distributed inference with uncertain models, which is significant for applications where agents learn likelihood models from limited data, impacting researchers and practitioners in distributed learning.
This paper addresses distributed classification where networked agents identify a target class from observations, specifically when agents have uncertain statistical models due to limited data. The authors propose a framework that allows agents' beliefs to converge to the same result as a centralized approach, even with parametric uncertainty, and explore extensions to non-parametric models.
This paper studies the problem of distributed classification with a network of heterogeneous agents. The agents seek to jointly identify the underlying target class that best describes a sequence of observations. The problem is first abstracted to a hypothesis-testing framework, where we assume that the agents seek to agree on the hypothesis (target class) that best matches the distribution of observations. Non-Bayesian social learning theory provides a framework that solves this problem in an efficient manner by allowing the agents to sequentially communicate and update their beliefs for each hypothesis over the network. Most existing approaches assume that agents have access to exact statistical models for each hypothesis. However, in many practical applications, agents learn the likelihood models based on limited data, which induces uncertainty in the likelihood function parameters. In this work, we build upon the concept of uncertain models to incorporate the agents' uncertainty in the likelihoods by identifying a broad set of parametric distribution that allows the agents' beliefs to converge to the same result as a centralized approach. Furthermore, we empirically explore extensions to non-parametric models to provide a generalized framework of uncertain models in non-Bayesian social learning.