ARTIFICIAL INTELLIGENCE
CSCI-UA 0472 · SPRING 2024 · COURANT INSTITUTE OF MATHEMATICAL SCIENCES
INSTRUCTOR | Alfredo Canziani, Ernest Davis |
LECTURES | Monday & Wednesday 11:00 – 12:15, Zoom |
MATERIAL | 2024 repo |
Context
It’s January 2024. I come back from the Winter Break and… I’m promoted to full-teaching faculty with new duties, including teaching an undergraduate Artificial Intelligence (AI) course here at NYU. We merged two offerings of this course and split the content among the two instructors. Ernie covered the Knowledge-Based AI part in the first half of the semester, while I took care of the Learning-Based AI & Natural Language Processing (NLP) bit. This second half of the course is what I’m releasing to the public. Working on this course and teaching my Deep Learning one took all my time. Therefore, I made no progress on the Energy-Based Deep Learning book this Spring semester.
Lectures
Legend: 🖥 slides, 📝 notes, 📓 Jupyter notebook, 🎥 YouTube video.
- Course first part recap, Naïve Bayes intro 🖥 🖥 🎥
- Discrete probability recap, Naïve Bayes classification 🖥 🖥 📝 🎥
- RN, chapter 12 – Quantifying uncertainty
- Naïve Bayes parameters estimation and Laplace smoothing 🖥 🎥
- Binary classifier evaluation, binary Perceptron 🖥 🖥 🎥
- Multiclass perceptron, binary and multiclass logistic regression 🖥 🖥 🎥
- Optimisation (gradient ascent) 🖥 🎥
- Statistical natural language processing 🖥
- RN, chapter 23 – Natural language processing
Sections 23.1.0–23.1.4, 23.1.7, 23.6 - JM, chapter 3 – N-gram language model
Sections 3.0–3.6
- RN, chapter 23 – Natural language processing
- Digit captioning 🖥
- Read programming assignment 4
- Classification with neural nets 🖥
- Language sampling and neural NLP 🖥
- JM, chapter 3 – N-gram language model
Section 3.4 - JM, chapter 6 – Vector semantics and embeddings
Sections 6.0, 6.3, 6.4, 6.9–6.11 - RN, chapter 23 – Natural language processing
Section 23.1.5 - RN, chapter 24 – Deep learning for NLP
Sections 24.0, 24.1, 24.2.0
- JM, chapter 3 – N-gram language model
- Convolutional nets + NB 🖥
- PyTorch convnet tutorial notebook. 📓
- Recurrent nets + NB 🖥
- PyTorch recurrent net tutorial notebook. 📓
- Recurrent nets for NLP and attention 🖥
- RN, chapter 24 – Deep learning for NLP
Sections 24.2–24.6
- RN, chapter 24 – Deep learning for NLP
Suggested readings
- RN: Russell & Norvig – Artificial Intelligence, 4th edition
- JM: Jurafsky & Martin – Speech and Language Processing, 3rd draft
- Books recommended during episode 13 (last class)
- François Fleuret – The Little Book of Deep Learning
- Sebastian Raschka