I'm trying to implement a soft actor-critic algorithm for financial data (stock prices), but I have trouble with losses: no matter what combination of hyper-parameters I enter, they are not converging, and basically it caused bad reward return as well. It sounds like the agent is not learning at all.
I already tried to tune some hyperparameters (learning rate for each network + number of hidden layers), but I always get similar results. The two plots below represent the losses of my policy and one of the value functions during the last episode of training.
My question is, would it be related to the data itself (nature of data) or is it something related to the logic of the code?