I implemented baseline predictors model (like it is told in Recommender systems handbook pp 148-149).
b_ui = mu + b_i + b_u
where mu is overall average rating and b_i, b_u are parameters. I calculate them with decoupling method, that uses parameters lambda_2, lambda_3. I firstly set parameters static 25 and 10 respectively, but it gives not enough accuracy. Can you tell how can I improve parameters using learning data?