Skip to content

comeh/DeepLearningForMDPs

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

DeepLearningForMDPs

Some deep learning-based algorithms are introduced and theoretically studied in DeepLearningForMDPs_theoreticalPart to solve Markovian Decisions Processes (MDPs). These latter are tested and compared on many numerical applications, and the results are available on DeepLearningForMDPs_applicationsPart.

  1. Some codes used for the tests presented in DeepLearningForMDPs_applicationsPart are available in this repertory:
  • slpde_HybridNow.py is the code, written in Python and TensorFlow, for the ClassifHybrid algorithm used in the Semi-Linear PDE example.
  • sgm_ClassifHybrid.py is the code, written in Python and TensorFlow, for the ClassifHybrid algorithm used in the Smart Grid Management example.
  • sgm_Qknn.jl is the code, written in Julia, for the Qknn algorithm used in the Smart Grid Management example.
  1. Decisions.mp4 is a video of the Qknn estimated optimal decisions to take w.r.t. time for the Smart Grid Management example. The terminal time was set to N=200.

Releases

No releases published

Packages

No packages published