[KongchangAI]
Concept概率逻辑策略梯度

PLPG

A neuro-symbolic reinforcement learning method that uses differentiable symbolic supervision to make the policy satisfy logical constraints during training, suited to discrete action spaces.

Timeline (last 90 days)

Nov 13

Probabilistic Logic Policy Gradient (PLPG) assumes that the action space is finite and discrete, which limits its application to continuous action spaces.

Unverified50%
Nov 13

A study proposed a PLPG-inspired continuous configuration method that uses reparameterized Monte Carlo sampling to estimate the probability that an action satisfies the investment mandate and adds it to the training objective.

Unverified50%

All Facts (2)

Source Articles