#Code by Edina Rosta; Simian Xing; Tiejun Wei; Yubing Lu; #Authors listed alphabetically and all authors contributed equally in this repo. #2023 June
this repoo contains a serie of python/matlab scripts, constructing simple 1D/2D systems. manipulating the FES, adding bias and calculating the mfpts. the general goal is to find optim way to add the bias(or set of bias) to minimize the mfpt of the system from starting state to the end state.
In the RL folders, a simple DQN agent is constructed to interact with the environment. the reward function is defined as the mfpt of the system. the state is defined as the current position of the system. the action is defined as the bias added to the system. the Q-table is updated by the Q-learning algorithm.
Unfinished work: Use the decision transformer for the 2D model.