This is the code for the version presented at NeurIPS. Compared to the JMLR version (master), it misses the improved version of SPDL and also uses a different agent architecture in the point mass environment.
This is the code for the version presented at NeurIPS. Compared to the JMLR version (master), it misses the improved version of SPDL and also uses a different agent architecture in the point mass environment.