ikostrikov / pytorch-trpo

PyTorch implementation of Trust Region Policy Optimization

ikostrikov/pytorch-trpo Issues

It seems that the importance sampling code part is wrong.
Updated 5 months ago1
what is shs?
Updated 2 years ago
What is get_kl() doing in main.py?
Closed 7 years ago12
Bootstrapping the value function?
Updated 3 years ago1
Idon‘t konw what the “neggdotstepdir” for ，Thanks ！！！
Updated 4 years ago
Object oriented
Updated 5 years ago
What and When to send on the GPU?
Closed 5 years ago1
Main.py line128
Closed 5 years ago
dose the linesearch method conflict with a "trust region" policy gradient algorithm?
Updated 5 years ago1
compute the Fisher-Vector Producy
Closed 6 years ago1
The step of t is not necessary in main.py
Updated 6 years ago
Is the get_kl() function correct?
Closed 6 years ago
other env
Closed 6 years ago1
what does volatile=True for?
Closed 6 years ago1
The get_kl function returns 0 always.
Closed 6 years ago1
doc?
Closed 6 years ago4
have you verified the code's correctness?
Closed 7 years ago1
How to modify the code for discrete actions?
Closed 7 years ago3
Have you tested it on the Atari games?
Closed 7 years ago6
Use pytorch 0.2.0?
Closed 7 years ago1