Machine Learning 2022 Spring
News
21
Feb
Github
課程Github: Github
Syllabus
Date | Topic | Preparation - zh | Preparation - en | Class Material | Extra Material |
---|---|---|---|---|---|
2/18 | Lecture 1:Introduction of Deep Learning |
影片 1
影片 2 |
Video 1
Video 2 |
2022: Introduction Week 1 Pytorch 1 Pytorch 2 Colab Rules: ppt / pdf Class intro: ppt / pdf Pytorch Tutorial 1: pdf Pytorch Tutorial 2: pdf Colab Tutorial: pdf Environment Setup pdf |
Introduction of Deep Learning
Backpropagation Predicting Pokémon CP Pokemon classification Logistic Regression |
2/25 | Lecture 2:What to do if my network fails to train |
影片 1
影片 2 影片 3 影片 4 影片 5 |
Video 1
Video 2 Video 3 Video 4 Video 5 |
2022: Basic theory Week 2 Theory: ppt / pdf |
Gradient Descent (Demo by AOE)
Beyond Adam (part 1) Beyond Adam (part 2) |
3/04 | Lecture 3:Image as input |
影片
|
Video
|
2022: Why deep/ Validation Week 3 - Validation Week 3 - Why Deep Validation: ppt / pdf Why Deep: ppt / pdf |
Spatial Transformer Layer
|
3/11 | Lecture 4:Sequence as input |
影片 1
影片 2 |
Video 1
Video 2 |
2022: 科技部 X Talent 課堂演講 |
RNN (part 1)
RNN (part 2) GNN (part 1) GNN (part 2) |
3/18 | Lecture 5:Sequence to sequence |
影片 1
影片 2 影片 3 |
Video 1
Video 2 Video 3 |
2022: Self-attention Week 5 xformer: ppt / pdf |
NAT model
Pointer network |
3/25 | Lecture 6:Generation |
影片 1
影片 2 影片 3 影片 4 |
Video 1
Video 2 Video 3 video 4 |
2022: Privacy for ML (吳沛遠老師授課) 2021: GAN: ppt / pdf |
Theory of GAN (part 1)
Theory of GAN (part 2) Theory of GAN (part 3) VAE FLOW-based Model |
4/01 | Recent Advance of Self-supervised learning for NLP |
影片 1
影片 2 影片 3 影片 4 |
Video 1
Video 2 Video 3 |
2022: Recent Advance of Self-supervised learning for NLP (姜成翰助教授課) Week 7 2021: BERT: ppt / pdf |
BERT (part 1)
BERT (part 2) GPT-3 |
4/08 | 期中考週不上課 | ||||
4/15 | Lecture 7:Self-supervised learning for Speech and Image |
2022:
SSL for Speech and Image: video ppt / pdf |
|||
4/22 | Lecture 8:Auto-encoder/ Anomaly Detection |
影片 1
影片 2 影片 3 影片 4 影片 5 影片 6 影片 7 影片 8 |
Video 1
Video 2 |
2022: 訊連玩美課堂演講 2021: Auto-Encoder: ppt / pdf |
PCA
t-SNE |
4/29 | Lecture 9:Explainable AI |
影片 1
影片 2 |
Video 1
Video 2 |
2022: Adversarial Attack for NLP PART:1 (姜成翰助教授課) Week 11 2021: Explainable AI: ppt / pdf |
|
5/06 | Lecture 10:Attack |
影片 1
影片 2 |
Video 1
Video 2 |
2022: Adversarial Attack for NLP PART:2 (姜成翰助教授課) Week 12 2021: Adversarial Attack: ppt / pdf |
|
5/13 | Lecture 11:Adaptation | 影片 1 | Video 2 | 2022: More about self-supervised learning: video ppt / pdf 2021: Adaptation: ppt / pdf |
|
5/20 | Lecture 12:Reinforcement Learning |
影片 1
影片 2 影片 3 影片 4 影片 5 |
Video 1
Video 2 Video 3 Video 4 Video 5 |
2022: Quantum Machine Learning (鄭皓中教授授課) 2021: DRL: ppt / pdf |
|
5/27 | Lecture 13:Network Compression |
影片 1
影片 2 |
Video 1
Video 2 |
2021: Network Compression: ppt / pdf |
PPO
Q-learning (part 1) Q-learning (part 2) |
6/03 | Lecture 14:Life-long
Learning
|
影片 1
影片 2 影片 3 |
Video 1
Video 2 |
2022: 期末考週不上課 2021: Life-long Learning: ppt / pdf |
|
6/10 | Lecture 15:Meta Learning |
影片 1
影片 2 |
Video 1
Video 2 |
2022: More about Meta Learning Meta ppt / pdf |
MAML (part 1)
MAML (part 2) MAML (part 3) MAML (part 4) MAML (part 5) MAML (part 6) MAML (part 7) MAML (part 8) MAML (part 9) Learnable Gradient Descent (pat 1) Learnable Gradient Descent (pat 2) Learnable Gradient Descent (pat 3) Metric-based (part 1) Metric-based (part 2) Metric-based (part 3) Train+Test as RNN |
Homework
The information here is tentative and subject to change. Please read the requirement of each homework before deadline.
- 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 will be subject to a 10% penalty if you violate any of the above rules.
- Prof. Lee & TAs preserve the rights to change the rules & grades.
Contact
Email: mlta-2022-spring@googlegroups.com
* 請同學來信務必依照正確寄信格式,並請善用NTU Cool討論區