tahamajs/DRL/homeworks - Storage Bucket

homeworks/ – Primary Homework Track

This folder contains the core homework sequence (HW1–HW14) plus supporting material. Each homework directory has a detailed README and notebooks/code for that unit.

Homework index

• HW1 – warm-up with classic control tasks; notebooks in code/ and base_code/, solutions in answer/.
• HW2_Value_Based_Methods – tabular and deep value-based RL.
• HW3_Policy_Gradients – REINFORCE and baseline techniques.
• HW4_Actor_Critic – actor–critic variants and advantage estimation.
• HW5_Model_Based – model learning and planning.
• HW6_Advanced_Topics – mixed advanced themes (check folder README).
• HW7_Deep_RL – larger-scale deep RL experiments.
• HW8_Exploration_Methods – intrinsic motivation and exploration bonuses.
• HW9_Advanced_Algorithms – performance/stability focused methods.
• HW10_Multi_Agent – cooperative/competitive MARL; includes base code and multiple tasks.
• HW11_Meta_Learning – fast adaptation/meta-RL setups.
• HW12_Hierarchical_RL – options/skills and temporal abstraction.
• HW13_Offline_RL – batch/offline policy learning.
• HW14_Safe_RL – constraints, safety layers, risk-sensitive training.

Supporting folders

• Weekly_Materials/ – small weekly practice exercises.
• Tasks_Special_Projects/ – project-style tasks with their own instructions.
• archive/ – past-term homework sets (e.g., sp23, sp24) with their own READMEs.
• Poster-Proposal-Template/ – template for course project proposals/posters.

How to run

1. Open the homework's README for dependencies.
2. Create a fresh virtual environment and install that homework's requirements.txt.
3. Run the provided notebook or script; submit outputs as directed in the homework README.