1
vote
Can action be dominated by state features in actor-critic algorithms?
It is certainly possible for the state features to dominate the action features in the critic.
There are several strategies you can use:
Replace the action features with a high dimensional learned ...
Only top scored, non community-wiki answers of a minimum length are eligible
Related Tags
td3 × 8reinforcement-learning × 6
deep-rl × 6
soft-actor-critic × 3
actor-critic-methods × 2
machine-learning × 1
ai-design × 1
hyper-parameters × 1
ddpg × 1
reward-functions × 1
action-spaces × 1
continuous-action-spaces × 1
javascript × 1
continuous-state-spaces × 1
activation-functions × 1