RFLMDA: A Novel Reinforcement Learning-Based Computational Model for Human MicroRNA-Disease Association Prediction

. 2021 Dec 5;11(12):1835. doi: 10.3390/biom11121835

Algorithm1: Pseudocode for RFLMDA algorithm.

Require: Action space A, state space S, reward value R, sub-models CMF, NRLMF and LapRLS.
Ensure: The predicted results of

F^{*};

1: Processing the dataset and training sub-models, namely CMF, NRLMF and LapRLS, respectively;
2: Calculation of the weights for models

F_{1}, F_{2} a n d F_{3}

via Pseudocode for Q-learning algorithm, respectively;
3: Combining

F_{1}, F_{2}, F_{3}

and S (a, b, c)

by F^{*} = a * F_{1} + b * F_{2} + c * F_{3}