python (65.1k questions)

javascript (44.2k questions)

reactjs (22.7k questions)

java (20.8k questions)

c# (17.4k questions)

html (16.3k questions)

r (13.7k questions)

android (12.9k questions)

Questions - policy-gradient-descent

Parallel environments in Pong keep ending up in the same state despite random actions being taken

Hi I am trying to use the SubprocVecEnv to run 8 parallel Pong environment instances. I tried testing the state transitions using random actions but after 15 steps (with random left or right actions),...
test-img

Swami

reinforcement-learning

openai-gym

pong

policy-gradient-descent

Votes: 0

Answers: 1

Latest Answer

Ok, figured it out. I was using the same seed for all the envs. envs=[make_env(env_name,42) for _ in range(8)] should be changed to envs=[make_env(env_name,i) for i in range(8)] #Seed as some functio...
test-img

Swami

python policy gradient reinforcement learning with continous action space is not working

i am trying to learn an agent to navigate to a target in my custom environment. The agent is learning with a neural net (2 hidden Dense layer, one dropout and one output layer of dimension 4). As inpu...
test-img

Viktoria

python

navigation

reinforcement-learning

montecarlo

policy-gradient-descent

Votes: 0

Answers: 1

Latest Answer

500 iterations are really really low for reinforcement learning problem, especially when neural networks are involved. Try increasing it (by a lot - at least 100.000 iterations), and also try to look ...
test-img

Raoul Raftopoulos

Action masking for continuous action space in reinforcement learning

Is there a way to model action masking for continuous action spaces? I want to model economic problems with reinforcement learning. These problems often have continuous action and state spaces. In add...
test-img

matthias

reinforcement-learning

openai-gym

policy-gradient-descent

sac

Votes: 0

Answers: 1

Latest Answer

I think you are on the right track. I've looked into masked actions and found two possible approaches: give a negative reward when trying to take an invalid action (without letting the environment evo...
test-img

Raoul Raftopoulos

Posts

Questions

Blogs

Jobs