Books
- AIMA - Artificial Intelligence: A Modern Approach by Stuart Russell, 4th edition, 2021. Also available at aima.cs.berkeley.edu. This book is required.
- GERON - Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow, 3rd Edition, 2022. Free for NJIT and NYU students. Very useful for those new to Python. For PyTorch users, see the open-source Dive into Deep Learning book.
- DL - Deep Learning. This book provides the necessary depth for statistical learning concepts in this course.
Planned Schedule
Part I: 2D Perception and Machine Learning
| Lecture | Topic | Description |
|---|---|---|
| 1 | Introduction to AI | Systems approach to AI. Overview of AI agents and course roadmap. Reading: AIMA Chapters 1 & 2 |
| 2 | Supervised Learning | Perception subsystem, reflexive agents, classification and regression with classical ML. Reading: AIMA Chapter 19 |
| 3 | Deep Neural Networks | From Perceptron to MLPs, SGD optimization, backpropagation fundamentals. Reading: AIMA Chapter 21, DL Chapter 6 |
| 4 | CNNs | Convolutional Neural Networks architecture and applications. Reading: DL Chapters 9 & 10, AIMA Chapter 25 |
| 5 | Scene Understanding | Object Detection, Semantic and Instance Segmentation. Reading: AIMA Chapter 25 |
| 6 | Probabilistic Models | Recursive state estimation, Dynamic Bayesian Networks, Kalman filters. Reading: AIMA Chapters 12, 13 & 14 |
Part II: Natural Language Processing
| Lecture | Topic | Description |
|---|---|---|
| 7 | NLP Fundamentals | Problem formulation and component mechanics. Reading: AIMA Chapter 23 |
| 8 | Language Modeling | RNN/LSTM architectures, attention mechanisms, Transformer framework. Reading: AIMA Chapter 24, DL Chapter 10 |
Part III: Reasoning and Planning without Interactions
| Lecture | Topic | Description |
|---|---|---|
| 9 | Knowledge Representation | Symbolic AI, propositional logic, theorem proving, Knowledge Base construction. Reading: AIMA Chapter 7 |
| 10 | Problem Solving and Search | Planning, A* algorithm, handling deterministic and stochastic environments. Reading: AIMA Chapters 3, 4 & 11 |
Part IV: Planning with Interactions - Reinforcement Learning
| Lecture | Topic | Description |
|---|---|---|
| 11 | MDPs and POMDPs | Sequential decision making, reward signals, policy computation. Reading: AIMA Chapters 16 & 17 |
| 12 | Reinforcement Learning | Bellman equations, policy iteration, REINFORCE algorithm with DNNs. Reading: AIMA Chapter 22 |
| 13 | Review | Final lecture before the exam |