Plan on grid world and must reach the end and avoid other block cars.
Use reinforcement learning to dispatch traffic police to deal with as many traffic accidents as possible.
Recommend the paper to experts according to the content of the paper and the interests and expertise of the experts.
里面有transformer的骚操作(作为伪encoder和decoder),以及针对2维卷积的attention。
Use multi-agent ReinForcement Learning on mobile crowd sensing.
This is a repository storing valuable paper and experiments implemented by SpartanBin about Reinforcement Learning.