Verification of AI

   verify geo modeling

CL#5 Reinforcement Learning (RL)

 自行決定是否做作業，做的話任選一題即可

 1. Use RL to train a robot to walk.

Must obey Physics law.

Start from zero knowledge.

2. Train the robot in the following scenario to cross the road.

scenario setting artifact

3. Use RL to train something you are interested in.

4. 市區交通模擬，如何控制燈號以增進運輸量 (RL)

    Birds' eye view and follow mode

 (credit: 與專題生共同創作)

Reinforcement Learning

 go A*

• RL (Reinforcement Learning)

Reinforcement Learning (slides)

A* for AI Optimization

 Ａ＊  search

• A* as heuristic (guidance)

• Optimize circuit design using A*

• AI is the new dev environment

• AI is the new UI.

Scientific Animation

3d 物件與互動 (水與冰)

Cat coat colors: 貓的花色遺傳模擬

AI for Scientific Discovery

AI weather prediction

e-Driving

share

 

Self-driving

UI change

AI weather prediction

Typhoon, share, artifact

artifact

CorrDiff video

 Perform Deep Research* and then animate an advanced AI tool (such as CorrDiff, Pangu, GraphCast ...) for weather prediction in Taiwan.

*

ChatGPT: 深入研究

Claude: Research

Grok: auto

Gemini 2.5 Flash: Deep Research