PPO Implementation

marked this issue as related to #2 (closed)

added Feature label

assigned to @dl337788

marked this issue as related to #17 (closed)

marked this issue as related to #18 (closed)

"If using a neural network architecture that shares parameters between the policy and value function, we must use a loss function that combines the policy surrogate and a value function error term." PPO p. 5

marked this issue as related to #21 (closed)

marked this issue as related to #22 (closed)

added In Progress label

marked this issue as related to #25 (closed)

changed milestone to %Official dates

Several performance optimizations are outlined and analyzed in Are Deep Policy Gradient AlgorithmsTruly Policy Gradient Algorithms? (Ilyas et al., 2018)

added In Review label and removed In Progress label

closed

removed In Review label

marked this issue as related to #35 (closed)

marked this issue as related to #36 (closed)

marked this issue as related to #33 (closed)

marked this issue as related to #31 (closed)

marked this issue as related to #53 (closed)