Financial Analysis 3

 2025 agent vs. market

heuristics 

heuristics vs. RL 

Follow up RL 3

 

artifact (share)

theme park (share)

premium 

PCB Layout Optimization  using RL (share)

RL for Circuit Layout Design

Advanced with stopping criteria, optimized 

F1 Racing by Mr. Chung

F1 Yas Marina by Lucas daylight, twilight

hotel staffing by Ms Lee

Pipeline by Mr. Hunter

RL for Preventive Maintenance

HW#13 Reinforcement Learning (RL) 3

  自行決定是否做作業，做的話任選一題即可

1. 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)

給高中生的導讀：看懂中壢交通模擬

sample, share

 2. Use RL to train Submarine to avoid unmanned torpedo from attaching and attacking

RL

share (artifact)

hi accomp share (artifact)

3. Use RL to optimize a Manufacturing Process overcoming a failure,  artifact (share) (inspired by Edward Chang)

  Job Shop

artifact (share)

4. Use RL to train inverted pendulum

artifact (share)

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

scenario setting artifact

artifact (share)

6. Use RL to train submarine maneuvers

Destroyer vs Submarine, share

RL artifact (share), h. accomplished, tactical

7. Use TL to train the AI manager of  Supermarket (share)

8. Use RL to train F1 racing

artifact (share)

9. Use RL to train TESLA FSD.

Financial analysis

TESLA  simple (share)

comprehensive (share)

Stock Price Projection TSMC, NVIDIA, TESLA

DD TSMC, TESLA

DCF TESLA

DCF-based Stock Price Projection

treasury bonds vs. fixed deposits

instruments vs. Taiwan real estate investment returns analysis

Taiwan fixed deposits

Taiwan youths and GDP growth, Asia

Follow up RL 2

 

• Traffic control by Mr. Chung
• F1 racing (share)

• Train the robot to cross the road.
• Supermarket (share)
• RL for Submarine Tactical Operation
• RL for Submarine Maneuvers (hellscape)

 

artifact (share)

theme park (share)

premium 

PCB Layout Optimization  using RL (share)

RL for Circuit Layout Design

Advanced with stopping criteria, optimized 

F1 Racing by Mr. Chung

F1 Yas Marina by Lucas daylight, twilight

hotel staffing by Ms Lee

Pipeline by Mr. Hunter

RL for Preventive Maintenance

• 
  

Train the robot to cross the road.

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

scenario setting artifact

artifact (share)

Robot Walk RL

 Robot Walk RL Gemini 3 Pro

Robot Walk RL, Opus 4.5 (share)

Simulating BLUE LED

  Blue LED

Simulating Laser

 雷射(2d 模擬 energy pumping)

share

雷射(3d 模擬) 使用3d量子場

Highly accomplished

share (He-Ne)

Space-Time of AI

  verify geo modeling

HW#12 Reinforcement Learning (RL) 2

 自行決定是否做作業，做的話任選一題即可

 2. Use RL to train Submarine to avoid unmanned torpedo from attaching and attacking

RL

share (artifact)

hi accomp share (artifact)

3. Use RL to optimize a Manufacturing Process overcoming a failure,  artifact (share) (inspired by Edward Chang)

  Job Shop

artifact (share)

4. Use RL to train inverted pendulum

artifact (share)

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

scenario setting artifact

artifact (share)

6. Use RL to train submarine maneuvers

Destroyer vs Submarine, share

RL artifact (share), h. accomplished, tactical

Job Shop RL

Visualization of Manufacturing Process,  simulation of pipelines (inspired by Edward Chang)

 

  Job Shop

artifact (share)

RL (Reinforcement Learning)

• slides
• 
  
• Inverted Pendulum (RL), Opus 4.6 (share)
•  drone training, terrain (share)
• 
  

• Train the robot to cross the road.
• Supermarket (share)
• 
  

• RL for ChungLi Traffic 400 tiers
• 
  

• RL for Submarine Tactical Operation
• RL for Submarine Maneuvers (Hellscape)
• drone training (share)

Protein Folding artifact  html

highly accomplished (share)

highly accomplish it artifact improved on student work

2 highly accomplish it artifact improved on student work

Inverted Pendulum (RL)

 

artifact (share)

RL for Submarine Tactical Operation

 

Destroyer vs Submarine, share

RL artifact (share), h. accomplished, tactical

RL for Submarine Maneuvers (Hellscape)

 Use RL to train Submarine to avoid unmanned torpedo from attaching and attacking

share (artifact)

hi accomp share (artifact)

simplified by Mr. M. Chang

RL

Also

evade mines using GA

RL for ChungLi Traffic 400 tiers

 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)

給高中生的導讀：看懂中壢交通模擬

sample, share

🚦 RL Traffic Control — Train, Save, Compare

Agents persist across sessions • Continue training anytime

🎮 Training (1747 episodes)

1x5x10x20x

1747

Episodes

5%

Exploration

1358079

Q-Table

-303.5

Avg R (10)

Learning Curve

🚦 RL Traffic Control — Train, Save, Compare

Agents persist across sessions • Continue training anytime

📊 Comparison Complete

Metric

❌ Untrained

✅ Trained (1688 eps)

Total Reward

-359.8

-158.8(+56%)

Throughput

17

21

Avg Wait

44.3

29.2

✓ Learning Verified! Trained agent outperformed random.