The main point of this paper is to provide an affirmative answer through exploiting reinforcement learning (RL) in artifcial intelligence (AI) for eliminating herding without any external control in complex resource allocation systems. In particular, we demonstrate that, when agents are empowered with RL (e.g., the popular Q-learning algorithm in AI) in that they get familiar with the unknown game environment gradually and attempt to deliver the optimal actions to maximize the payoff, herding can effectively be eliminated. Furthermore, computations reveal the striking phenomenon that, regardless of the initial state, the system evolves persistently and relentlessly toward the optimal state in which all resources are used efficiently. However, the evolution process is not without interruptions: there are large fluctuations that occur but only intermittently in time. The statistical distribution of the time between two successive fluctuating events is found to depend on the parity of the evolution, i.e., whether the number of time steps in between is odd or even. We develop a physical analysis and derive mean-field equations to gain an understanding of these phenomena. Since AI is becoming increasingly widespread, we expect our RL empowered minority game system to have broad applications.

Supply-demand processes take place on a large variety of real-world networked systems ranging from power grids and the internet to social networking and urban systems. In a modern infrastructure, supply-demand systems are constantly expanding, leading to constant increase in load requirement for resources and consequently, to problems such as low efficiency, resource scarcity, and partial system failures. Under certain conditions global catastrophe on the scale of the whole system can occur through the dynamical process of cascading failures. We investigate optimization and resilience of time-varying supply-demand systems by constructing network models of such systems, where resources are transported from the supplier sites to users through various links. Here by optimization we mean minimization of the maximum load on links, and system resilience can be characterized using the cascading failure size of users who fail to connect with suppliers. We consider two representative classes of supply schemes: load driven supply and fix fraction supply. Our findings are:(1) optimized systems are more robust since relatively smaller cascading failures occur when triggered by external perturbation to the links; (2) a large fraction of links can be free of load if resources are directed to transport through the shortest paths; (3) redundant links in the performance of the system can help to reroute the traffic but may undesirably transmit and enlarge the failure size of the system; (4) the patterns of cascading failures depend strongly upon the capacity of links; (5) the specific location of the trigger determines the specific route of cascading failure, but has little effect on the final cascading size; (6) system expansion typically reduces the efficiency; and (7) when the locations of the suppliers are optimized over a long expanding period, fewer suppliers are required. These results hold for heterogeneous networks in general, providing insights into designing optimal and resilient complex supply-demand systems that expand constantly in time.challenging, attracting continuous interest of researchers from various disciplines such as business, economics, systems engineering, computer science, geography, and even biology [3][4][5][6][7][8][9][10][11].In this paper, we investigate optimization and resilience of complex demand-supply networks from the dynamical point of view, motivated by the fact that, in general, rapid expansion of any networked system will inevitably affect the various dynamical processes that it supports. For a supply-demand network, expansion can lead to increasing load requirement for resources, causing problems such as low efficiency, resource scarcity, and small and large scale failures. Of particular interest is the dynamical process of cascading failures, which has been studied extensively in the past but mostly for static networks [12][13][14][15][16][17][18][19][20][21][22][23][24][25][26]. There were also previous works on dynamical processes on time-varying networks [27], in specific contexts su...