Machine Learning Advanced Pro

Policy Gradients & Actor-Critic

Directly optimizing the policy — REINFORCE, baselines, advantage, and A2C

30 min read 19 views

Learn the policy gradient theorem intuitively, REINFORCE as Monte Carlo policy gradient, why it has high variance, how a baseline and the advantage function reduce it without introducing bias, and how Advantage Actor-Critic (A2C) combines a policy (actor) with a learned value function (critic) into a stable, online training loop.

Part of the Reinforcement Learning & Long-term Optimization path →

Preview Subscribe to unlock

Practice questions (5)

Compute a REINFORCE Update With and Without a Baseline

Advanced · Free

View →
Choose an Advantage Estimator Under a Bias/Variance Constraint

Advanced

View →
Debug a Collapsing Shared-Trunk Actor-Critic

Advanced

View →
Evaluate a Proposal to Add a Replay Buffer to A2C

Advanced

View →
Design a Policy Head for a Continuous Pricing Action

Advanced

View →

Quizzes (1)

Policy Gradients & Actor-Critic Quiz

0 attempts · 10 min · Pass 70% · Advanced

Take quiz →