arXiv subjects

Jemin George

Publications and source records attributed to Jemin George.

25 records · Page 2Linked to original sources

arXiv2019

Distributed Deep Learning with Event-Triggered Communication

We develop a Distributed Event-Triggered Stochastic GRAdient Descent (DETSGRAD) algorithm for solving non-convex optimization problems typically encountered in distributed deep learning. We propose a novel communication triggering mechanism that would allow the networked agents to update their model parameters aperiodically and provide sufficient conditions on the algorithm step-sizes that guarantee the asymptotic mean-square convergence. The algorithm is applied to a distributed supervised-learning problem, in which a set of networked agents collaboratively train their individual neural networks to recognize handwritten digits in images, while aperiodically sharing the model parameters with their one-hop neighbors. Results indicate that all agents report similar performance that is also comparable to the performance of a centrally trained neural network, while the event-triggered communication provides significant reduction in inter-agent communication. Results also show that the proposed algorithm allows the individual agents to recognize the digits even though the training data corresponding to all the digits are not locally available to each agent.

math.OC↗

arXiv2019

Distributed Stochastic Gradient Method for Non-Convex Problems with Applications in Supervised Learning

We develop a distributed stochastic gradient descent algorithm for solving non-convex optimization problems under the assumption that the local objective functions are twice continuously differentiable with Lipschitz continuous gradients and Hessians. We provide sufficient conditions on step-sizes that guarantee the asymptotic mean-square convergence of the proposed algorithm. We apply the developed algorithm to a distributed supervised-learning problem, in which a set of networked agents collaboratively train their individual neural nets to recognize handwritten digits in images. Results indicate that all agents report similar performance that is also comparable to the performance of a centrally trained neural net. Numerical results also show that the proposed distributed algorithm allows the individual agents to recognize the digits even though the training data corresponding to all the digits is not locally available to each agent.

math.OC↗

arXiv2018

Distributed Robust Dynamic Average Consensus with Dynamic Event-Triggered Communication

This paper presents the formulation and analysis of a fully distributed dynamic event-triggered communication based robust dynamic average consensus algorithm. Dynamic average consensus problem involves a networked set of agents estimating the time-varying average of dynamic reference signals locally available to individual agents. We propose an asymptotically stable solution to the dynamic average consensus problem that is robust to network disruptions. Since this robust algorithm requires continuous communication among agents, we introduce a novel dynamic event-triggered communication scheme to reduce the overall inter-agent communications. It is shown that the event-triggered algorithm is asymptotically stable and free of Zeno behavior. Numerical simulations are provided to illustrate the effectiveness of the proposed algorithm.

math.OC↗

arXiv2018

Distributed Optimization for Second-Order Multi-Agent Systems with Dynamic Event-Triggered Communication

In this paper, we propose a fully distributed algorithm for second-order continuous-time multi-agent systems to solve the distributed optimization problem. The global objective function is a sum of private cost functions associated with the individual agents and the interaction between agents is described by a weighted undirected graph. We show the exponential convergence of the proposed algorithm if the underlying graph is connected, each private cost function is locally gradient-Lipschitz-continuous, and the global objective function is restricted strongly convex with respect to the global minimizer. Moreover, to reduce the overall need of communication, we then propose a dynamic event-triggered communication mechanism that is free of Zeno behavior. It is shown that the exponential convergence is achieved if the private cost functions are also globally gradient-Lipschitz-continuous. Numerical simulations are provided to illustrate the effectiveness of the theoretical results.

math.OC↗

arXiv2018

Distributed Fault Detection and Accommodation in Dynamic Average Consensus

This paper presents the formulation of fault detection and accommodation schemes for a network of autonomous agents running internal model-based dynamic average consensus algorithms. We focus on two types of consensus algorithms, one that is internally stable but non-robust to initial conditions and one that is robust to initial conditions but not internally stable. For each consensus algorithm, a fault detection filter based on the unknown input observer scheme is developed for precisely estimating the communication faults that occur on the network edges. We then propose a fault remediation scheme so that the agents could reach average consensus even in the presence of communication faults.

math.OC↗

arXiv2018

Distributed Maximum Likelihood using Dynamic Average Consensus

This paper presents the formulation and analysis of a novel distributed maximum likelihood algorithm that utilizes a first-order optimization scheme. The proposed approach utilizes a static average consensus algorithm to reach agreement on the initial condition to the iterative optimization scheme and a dynamic average consensus algorithm to reach agreement on the gradient direction. The current distributed algorithm is guaranteed to exponentially recover the performance of the centralized algorithm. Though the current formulation focuses on maximum likelihood algorithm built on first-order methods, it can be easily extended to higher order methods. Numerical simulations validate the theoretical contributions of the paper.

math.OC↗

arXiv2016

Strategic Seeding of Rival Opinions

We present a network influence game that models players strategically seeding the opinions of nodes embedded in a social network. A social learning dynamic, whereby nodes repeatedly update their opinions to resemble those of their neighbors, spreads the seeded opinions through the network. After a fixed period of time, the dynamic halts and each player's utility is determined by the relative strength of the opinions held by each node in the network vis-a-vis the other players. We show that the existence of a pure Nash equilibrium cannot be guaranteed in general. However, if the dynamics are allowed to progress for a sufficient amount of time so that a consensus among all of the nodes is obtained, then the existence of a pure Nash equilibrium can be guaranteed. The computational complexity of finding a pure strategy best response is shown to be NP-complete, but can be efficiently approximated to within a (1 - 1/e) factor of optimal by a simple greedy algorithm.

cs.GT↗