How can I extract a trained RL Agent's network's weights and biases?

10 ビュー (過去 30 日間)
rakbar
rakbar 2020 年 3 月 26 日
コメント済み: 2024 年 1 月 5 日
How can I extract a trained RL Agent's network's weights and biases?
My network is:
statePath = [
imageInputLayer([numObservations 1 1], 'Normalization', 'none', 'Name', 'state')
fullyConnectedLayer(NumNeuron, 'Name', 'CriticStateFC1')
reluLayer('Name', 'CriticRelu1')
fullyConnectedLayer(NumNeuron, 'Name', 'CriticStateFC2')];
actionPath = [
imageInputLayer([1 1 1], 'Normalization', 'none', 'Name', 'action')
fullyConnectedLayer(NumNeuron, 'Name', 'CriticActionFC1')
reluLayer('Name', 'ActorRelu1')
fullyConnectedLayer(NumNeuron, 'Name', 'CriticActionFC2')];
commonPath = [
additionLayer(2,'Name', 'add')
reluLayer('Name','CriticCommonRelu')
fullyConnectedLayer(1, 'Name', 'output')];
criticNetwork = layerGraph(statePath);
criticNetwork = addLayers(criticNetwork, actionPath);
criticNetwork = addLayers(criticNetwork, commonPath);
criticNetwork = connectLayers(criticNetwork,'CriticStateFC2','add/in1');
criticNetwork = connectLayers(criticNetwork,'CriticActionFC2','add/in2');
% set some options for the critic
criticOpts = rlRepresentationOptions('LearnRate',learing_rate,...
'GradientThreshold',1);
% create the critic based on the network approximator
critic = rlQValueRepresentation(criticNetwork,obsInfo,actInfo,...
'Observation',{'state'},'Action',{'action'},criticOpts);
agent = rlDQNAgent(critic,agentOpts)
trainingStats = train(agent,env,trainOpts);
After training, I'd like to get the network's trained weights and biases.

採用された回答

Anh Tran
Anh Tran 2020 年 3 月 27 日
編集済み: Anh Tran 2020 年 3 月 27 日
You can get the parameters from the trained's critic representation for DQN agent. In MATLAB R2020a, see getLearnableParameters and getCritic functions (function name changes a bit since R2019b). You can follow similar steps to get the actor's parameters from actor-based agent like DDPG or PPO.
critic = getCritic(agent);
criticParams = getLearnableParameters(critic);
  6 件のコメント
Francisco Serra
Francisco Serra 2023 年 12 月 14 日
@rakbar @Dmitriy Ogureckiy have you found a way of getting the weights after each training episode?
轩
2024 年 1 月 5 日
@Francisco Serra I have the same need. I find a silly method: save the agent after each episode and use "getLearnableParameters" to print the parameter of each agent.

サインインしてコメントする。

その他の回答 (0 件)

Community Treasure Hunt

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

Start Hunting!

Translated by