Depthwise Convolution for Multi-Agent Communication With Enhanced Mean-Field Approximation.

Multi-Agent settings remain a fundamental challenge in the reinforcement learning (RL) domain due to the partial observability and the lack of accurate real-time interactions across agents. In this article, we propose a new method based on local communication learning to tackle the multi-agent RL (MARL) challenge within a large number of agents coexisting. First, we design a new communication protocol that exploits the ability of depthwise convolution to efficiently extract local relations and learn local communication between neighboring agents. To facilitate multi-agent coordination, we explicitly learn the effect of joint actions by taking the policies of neighboring agents as inputs. Second, we introduce the mean-field approximation into our method to reduce the scale of agent interactions. To more effectively coordinate behaviors of neighboring agents, we enhance the mean-field approximation by a supervised policy rectification network (PRN) for rectifying real-time agent interactions and by a learnable compensation term for correcting the approximation bias. The proposed method enables efficient coordination as well as outperforms several baseline approaches on the adaptive traffic signal control (ATSC) task and the StarCraft II multi-agent challenge (SMAC).

Effect of Nanosized NbC Precipitates on Hydrogen Diffusion in X80 Pipeline Steel.

In this paper, the effects of dispersed 3~10 nm NbC precipitates on hydrogen diffusion in X80 pipeline steel were investigated by means of high resolution transmission electron microscopy (HRTEM), electrochemical hydrogen permeation, and thermal desorption spectroscopy (TDS). The relationship between hydrogen diffusion and temperature was determined for Nb-free X80 and 0.055 wt% Nb X80 steel. The temperature dividing reversible and irreversible traps was measured, and the quantity of hydrogen captured by different traps was calculated. Three types of hydrogen trap were designed and applied in the test, and the results revealed that irreversible hydrogen traps formed by nanosized and coherent NbC precipitates markedly hindered hydrogen diffusion, and prolonged breakthrough time in Nb-bearing X80 steel.