xiliu18 / cs294-112-solutions

My implementation of DeepRL algorithms through assignments of cs294-112 Berkeley 2018 Fall

cs294-112-solutions

My solutions to assignments of Deep Reinforcement Learning (cs294-112 Berkeley, 2018 Fall). In each assignment, the following reinforcement learning algorithms were implemented:

• Homework 1: Imitation Learning: Behavioral Cloning and DAgger.
• Homework 2: Policy Gradients: Policy Gradient with advantage-centering, reward-to-go, and state-dependent baseline.