Assignments for Berkeley CS 285: Deep Reinforcement Learning, Decision Making, and Control.

This is for 2023 fall assignment version

Author: Bowie Shi

Email: aoibosh@connect.hku.hk

Data: started from 2024-1-25

• homework 1 finished.
• homework 2 finished, experiment results satisfy requirements.
• homework 3 finished, one sac reinforment part experiment result is not satisfied.
• homework 4 finished, experiment results satisfy requirements.
• homework 5, to be done

If you find some mistakes, feel free to email me with your suggestion. I will be more than gratitude.

If you are working on homework 3 or 5, feel free to email me. I am finding groupmates to discuss the assignment.

reference (mostly 2020 version assignemt):

https://gitee.com/kin_zhang/drl-hwprogramm/blob/solution/hw2/solution.md

https://gitee.com/kin_zhang/drl-hwprogramm/blob/solution/hw2/cs285/agents/pg_agent.py

https://gitee.com/kin_zhang/drl-hwprogramm/blob/solution/hw2/cs285/policies/MLP_policy.py