GraphMIX
GraphMIX copied to clipboard
navid-naderi
Implementation code for GraphMIX: Graph Convolutional Value Decomposition in Multi-Agent Reinforcement Learning
Hello, I see that your article mentioned the permutation invariance in the past, I would like to ask if the algorithm can achieve a transferable MADRL through the permutation invariance,...
In paper, figure 5 GraphMIX (GCN) results are there. However, I could not able to find layers_gcn.py which contains "class GraphConvolution(nn.Module):" Kindly look into this. Regards Ajay N N
Metadata
Owner
Metadata
Implementation code for GraphMIX: Graph Convolutional Value Decomposition in Multi-Agent Reinforcement Learning