cho3/ReinforcementLearning.jl
Something that tries to tie together various online model-free methods into a vaguely extensible and sensible api
Jupyter Notebook
Issues
- 1
- 0
Break up policy.jl
#5 opened by cho3 - 0
How to handle different kinds of learning rate annealing and make it consistent with different update methods
#4 opened by cho3 - 1
Figure out exploration policies
#1 opened by cho3 - 0
Figure out how to elegantly handle feature functions that handle states and actions, and just states
#2 opened by cho3