Why my critic agent works with DQN, and it doesn't work with Q-learning or SARSA agents for the exactly same environment??

Question

Juan Vasconez 2021-5-13

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/829213-why-my-critic-agent-works-with-dqn-and-it-doesn-t-work-with-q-learning-or-sarsa-agents-for-the-exac

编辑： Juan Vasconez 2021-5-14

%Why my critic agent works with DQN, and it doesn't work with Q-learning or SARSA agents for the exactly same environment??

% AFTER USING The TRAIN fucntion with the rlQAgent I GET THIS ERROR

Error using rl.policy.AbstractPolicy/step (line 242)

Invalid input argument type or size such as observation, reward, isdone or loggedSignals.

Error in rl.env.MATLABEnvironment/simLoop (line 258)

action = step(policy,observation,reward,isdone);

Error in rl.env.MATLABEnvironment/simWithPolicyImpl (line 106)

[expcell{simCount},epinfo,siminfos{simCount}] = simLoop(env,policy,opts,simCount,usePCT);

Error in rl.env.AbstractEnv/simWithPolicy (line 83)

[experiences,varargout{1:(nargout-1)}] = simWithPolicyImpl(this,policy,opts,varargin{:});

Error in rl.task.SeriesTrainTask/runImpl (line 33)

[varargout{1},varargout{2}] = simWithPolicy(this.Env,this.Agent,simOpts);

Error in rl.task.Task/run (line 21)

[varargout{1:nargout}] = runImpl(this);

Error in rl.task.TaskSpec/internal_run (line 166)

[varargout{1:nargout}] = run(task);

Error in rl.task.TaskSpec/runDirect (line 170)

[this.Outputs{1:getNumOutputs(this)}] = internal_run(this);

Error in rl.task.TaskSpec/runScalarTask (line 194)

runDirect(this);

Error in rl.task.TaskSpec/run (line 69)

runScalarTask(task);

Error in rl.train.SeriesTrainer/run (line 24)

run(seriestaskspec);

Error in rl.train.TrainingManager/train (line 424)

run(trainer);

Error in rl.train.TrainingManager/run (line 215)

train(this);

Error in rl.agent.AbstractAgent/train (line 77)

TrainingStatistics = run(trainMgr);

Error in QNN_train_emg_RLmat_ToolboxV1 (line 132)

trainStats = train(agent,env,opt);

Caused by:

Error using rl.representation.rlAbstractRepresentation/validateInputData (line 509)

Input data must be a cell array of compatible dimensions with observation and action info specifications.

%-----------------------------------------------------------------------------------------------------------------------------

%There is an error in the function AbstractPolicy.m

% call the step implementation with the generated experience

%line 237% action = stepImpl(this,exp); %aqui esta el error

%I do not have this issue when using agent = rlDQNAgent(NeuralN);

I only have it when using -> agent = rlQAgent(NeuralN); OR agent = rlSARSAAgent(NeuralN);

It seems that the ExperienceBuffer instance is missing in the variable called "this" when I use agent = rlQAgent(NeuralN); instead of agent = rlDQNAgent(NeuralN);

Any suggestions??????

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Why my critic agent works with DQN, and it doesn't work with Q-learning or SARSA agents for the exactly same environment??

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

回答（0 个）

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

Why my critic agent works with DQN, and it doesn't work with Q-learning or SARSA agents for the exactly same environment??

0 个评论 显示 -2更早的评论隐藏 -2更早的评论

回答（0 个）

另请参阅

类别

标签

产品

版本

Community Treasure Hunt

0 个评论
显示 -2更早的评论隐藏 -2更早的评论