DQN

class DQN(model, gamma=None, lr=None)

Bases: parl.core.paddle.algorithm.Algorithm

__init__(model, gamma=None, lr=None)

DQN algorithm

Parameters:

• model (parl.Model) – forward neural network representing the Q function.
• gamma (float) – discounted factor for accumulative reward computation
• lr (float) – learning rate.

learn(obs, action, reward, next_obs, terminal)

update the Q function (self.model) with DQN algorithm

predict(obs)

use self.model (Q function) to predict the action values

sync_target()

assign the parameters of the training network to the target network