Description
作業內容
在本次作業當中，你們將可以實做幾項 Deep Reinforcement Learning 方法：
- Policy Gradient
- Actor-Critic
- 作業的實做環境為 OpenAI 的 gym 當中的 Lunar Lander。其餘實做細節請參考助教提供的範例程式。
範例展示
Policy Gradient 方法(
Actor-Critic 方法
範例結果
繳交項目及評分標準
- Python 程式碼 ( Submit on NTU COOL) 佔4分
- Action List ( Submit on JudgeBoi, 沒有private set, 自動選擇最高分)
繳交項目及評分標準
More on a “valid submission “:
agent在action list最後一個動作輸入之後，應該要輸出done。長度過長或過短的 action list都會被系統reject。
Bonus
- If you successfully get 10 pts:
○ Your code will be made public to students.
○ You can submit a report in PDF format briefly describing what you have done (in English, less than 100 words) for extra 0.5 pts.
○ Reports will also be made public to students.
○ Notice, we do not have private score, so omit it in the report.
- Report template
注意事項
- You should finish your homework on your own.
- You should NOT modify your prediction files manually.
- Do NOT share codes or prediction files with any living creatures.
- Do NOT use any approaches to submit your results more than 5 times a day.
- Do NOT search or use additional data or pre-trained models.
- Your final grade x 0.9 if you violate any of the above rules.
Prof. Lee & TAs preserve the rights to change
