Assistant Professor
PhD (University of Notre Dame, 2020)
Email: chuxuzhang[at]brandeis[dot]edu
COSI 165B: Deep Learning (Spring 2021, Brandeis)
Course Information
- Course Description: This course is an advanced level course for undergraduates (computer science major) and graduate students in Arts & Sciences. Due to its powerful capability and excellent performance in solving real-world problems, deep learning has become one of the most important machine learning techniques. This course covers the core methods and algorithms of deep learning techniques. Students learn models and algorithms of deep learning, apply deep learning tools, and work on related homework and course project to addresse real world problems. Topics include: machine learning basics, feed-forward neural networks, neural network optimization, deep learning tools, convolutional neural networks, recurrent neural networks, graph neural networks, attention networks, auto-encoder networks, generative networks, and others.
- Instructor: Chuxu Zhang (chuxuzhang@brandeis.edu)
- Teaching Assistant: Yue Han (hanyue@brandeis.edu)
- Lecture Time: Monday/Wednesday 4:00-5:30 PM
- Location: Zoom
- Instructor Office Hour: Tuesday 9-11 PM
- TA Office Hour: Friday 8-10 PM
- Forum: LATTE
Logistics
- Prerequisites: COSI 21a, MATH 8a, MATH 10a, (MATH 15a or MATH 22a), proficiency in Python programming, machine learning (preferred).
- Course materials:
Deep Learning, Ian Goodfellow and Yoshua Bengio and Aaron Courville, MIT Press
Dive into Deep Learning, Aston Zhang, Zachary C. Lipton, Mu Li, and Alexander J. Smola - Grading
Contributions: four programming homework (60% = 4 * 15%), one course project with presentation (40%: midterm proposal/presentation 10%, final presentation 10%, final report 20%).
Grades (4 points in Brandeis): A (A+) = 4.00; A- = 3.67; B+ = 3.33; B = 3.00; B- = 2.67; C+ = 2.33; C = 2.00; C- = 1.67; D+ = 1.33; D = 1.00; D- = 0.67 and E = 0.00
Grades (100 points): A (A+) = [90, 100]; A- = [87, 90); B+ = [84, 87); B = [81, 84); B- = [78, 81); C+ = [75, 78); C = [72, 75); C- = [69, 72); D+ = [66, 69); D = [63, 66); D- = [60, 63); and E = [0, 60) - Statement for Covid-19: online lecture and office hour through Zoom.
Content/Schedule
| Date | Topic | Content | Event |
| 2/1 | Introduction/Background | Introduction of Deep Learning | |
| 2/3 | Introduction/Background | Machine Learning Basics | |
| 2/8 | Feed-forward Neural Networks | Peceptron, MLP | |
| 2/10 | Feed-forward Neural Networks | Backpropagation | |
| 2/15 | No lecture | First Homework Release | |
| 2/17 | Pytorch | Pytorch Tutorial | |
| 2/22 | Neural Network Optimization | Regularization | |
| 2/24 | Neural Network Optimization | Optimization | |
| 3/1 | Convolutional Neural Network | Convolution, Pooling | First Homework Due |
| 3/3 | Convolutional Neural Network | Neuroscientific Basis, Basic CNN | |
| 3/8 | Convolutional Neural Network | Advanced CNN | |
| 3/10 | Recurrent Neural Network | RNN Basis | Second Homework Release |
| 3/15 | Recurrent Neural Network | Advanced RNN | |
| 3/17 | Attention | Attention, Transformer | |
| 3/22 | Auto-Encoder | Auto-Encoder | |
| 3/24 | DL for NLP | Word/Text Embedding | Second Homework Due |
| 3/29 | No Lecture | ||
| 3/31 | Project Midterm Presentation | Project Proposal Due | |
| 4/5 | Graph Neural Network | Graph Embedding Introduction | Third Homework Release |
| 4/7 | Graph Neural Network | Network Embedding | |
| 4/12 | Graph Neural Network | Graph Neural Network | |
| 4/14 | Graph Neural Network | GNN Applications | |
| 4/19 | Deep Generative Network | Variational Auto-encoder | Third Homework Due |
| 4/21 | Deep Generative Network | Generative Adversarial Network | Fourth Homework Release |
| 4/26 | Seminar | Recommender Systems | |
| 4/28 | Seminar | BERT | |
| 5/3 | Content Review | ||
| 5/5 | Project Final Presentation | 5/9 Fourth Homework Due 5/12 Project Report Due |