Switching Transition in a Resource Exchange Model on Graphs

TL;DR

This paper studies a resource exchange model on various random graphs, revealing a sharp transition between token-rich and token-empty states driven by a control parameter, with topology influencing the critical point.

Contribution

It introduces a nonequilibrium resource exchange model on complex networks and characterizes the topology-dependent switching transition between different steady states.

Findings

System exhibits a sharp switch-like transition in token distribution.

Most tokens condense on minimum degree nodes in non-regular graphs.

Critical threshold depends on graph topology.

Abstract

In this work, we investigate a simple nonequilibrium system with many interconnected, open subsystems, each exchanging a globally conserved resource with an external reserve. The system is represented by a random graph, where nodes represent the subsystems connected through edges. At each time step, a randomly selected node gains a token (i.e, a resource) from the reserve with probability (1-p) or loses a token to the reserve with probability p. When a node loses a token, its neighbors also lose a token each. This asymmetric token exchange breaks the detailed balance. We investigate the steady state behavior of our model for different types of random graphs: graphs without edges, regular graphs, Erd\H{o}s-R\'enyi, and Barab\'asi-Albert graphs. In all cases, the system exhibits a sharp, switch-like transition between a token-saturated state and an empty state. When the control parameter p is below a critical threshold, almost all tokens accumulate on the graph. Furthermore, in a non-regular graph, most tokens accumulate or condense on nodes of minimum degree. A slight increase in p beyond the threshold drains almost all the tokens from the graph. This switching transition results from the interplay between drift and the conservation of tokens. However, the position of the critical threshold and the behavior at the transition zone depend on graph topology.