Reinforcement Learning experience buffer length and parallelisation toolbox
显示 更早的评论
When parallelisation is used when training a DDPG agent with the following settings:
trainOpts.UseParallel = true;
trainOpts.ParallelizationOptions.Mode = 'async';
trainOpts.ParallelizationOptions.StepsUntilDataIsSent = -1;
trainOpts.ParallelizationOptions.DataToSendFromWorkers = 'Experiences';
Does the the parallel simulations have their own experience buffer? This could take up more memory hence I am hoping that only one experience buffer is stored to update the critic network.
From the documentations, it seems like there will only be one experience buffer as the experiences are sent back to the host.
采纳的回答
Emmanouil Tzorakoleftherakis
2020-12-3
编辑：Emmanouil Tzorakoleftherakis
2020-12-3
0 个投票
Hello,
There is one big experience buffer on the host, the size of which you determine as usual in your agent options. Each worker has a much smaller buffer to collect experiences until you reach "StepsUntilDataIsSent".
更多回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Deep Learning Toolbox 的更多信息
2020-12-2
2020-12-3
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
