How to interpret the learnableParameters (Reinforcement Learning Toolbox)?

3 ビュー (過去 30 日間)
black_cat
black_cat 2021 年 6 月 10 日
コメント済み: Shadaab Siddiqie 2021 年 6 月 15 日
I was wondering how to interpret the tables that one is receiving when using this line:
params = getLearnableParameters (critic);
Why are there 6 tables after q learning for example? I expected one.

採用された回答

black_cat
black_cat 2021 年 6 月 15 日
The values in these tables refer to the weights for the critic network.

その他の回答 (1 件)

Shadaab Siddiqie
Shadaab Siddiqie 2021 年 6 月 15 日
From my understanding you want to know why are there 6 values returned as the result of the getLearnableParameters function. I assume that you are running example in getLearnableParameters:
load('DoubleIntegDDPG.mat','agent')
critic = getCritic(agent);
params = getLearnableParameters(critic);
In this case since DoubleIntegDDPG.mat has 6 learnable parameters from the reinforcement learning policy (critic in your case) getLearnableParameters returns 6 values.
  2 件のコメント
black_cat
black_cat 2021 年 6 月 15 日
No. I figured it out: Those values in these 6 tables refer to the weights in my dnn. I appreciate your effort though.
Shadaab Siddiqie
Shadaab Siddiqie 2021 年 6 月 15 日
Great

サインインしてコメントする。

製品


リリース

R2021a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by