|
Sep 07, 2024
|
|
|
|
CSC 6621 - Applied Machine Learning4 lecture hours 0 lab hours 4 credits Course Description This course expands on the introduction to data science and machine learning in CSC 5610 with coverage of deep learning and applied artificial intelligence. Students will be introduced to deep learning, applications to complex data such as images and text, and appropriate libraries. Students will explore the paradigm of artificial intelligence through concepts such as agent-based frameworks, planning problems, algorithms such as reinforcement learning, and applications to playing games. Students will participate in a seminar-like experience where primary literature is read and presented. A final project will allow students to explore topics and applications beyond those presented in class. Students will reinforce their learning of machine learning algorithms with hands-on laboratory exercises for development of representative applications. (prereq: (CSC 5610 or CSC 2621) and (MTH 2130 or MTH 2340 or MTH 5810 ) or instructor consent) (quarter system prereq: CS 2300) Course Learning Outcomes Upon successful completion of this course, the student will be able to:
- Understand the fundamental concepts and application of supervised, unsupervised, semi-supervised, and reinforcement learning
- Understand the concepts of learning theory, (e.g., what is learnable, bias, variance, overfitting, curse of dimensionality, splitting data for evaluation of model predictions)
- Derive, visualize, apply, and interpret metrics and loss functions to assess the quality of model predictions and fit to data
- Describe the fundamentals of deep learning and what is required to be able to use it
- Compare various modern deep neural network architectures (e.g., convolutional neural networks (CNNs), encoder-decoders, generative adversarial networks (GANs), or transformers).
- Implement, train, and evaluate deep learning models using popular libraries
- Describe applications of deep learning to solving real-world problems
- Describe the components of the intelligent agent framework
- Apply agent and state-based frameworks to solve various problems such as game playing
- Provide examples of search problems and techniques for solving them
- Describe and apply algorithms such as value iteration and q-learning to identify optimal policies
- Describe the role of heuristics and describe the trade-offs among completeness, optimality, time complexity, and space complexity
- Read, interpret, and explain primary literature from the field of AI
- Present the results of ML and AI projects to technical and non-technical audiences
Prerequisites by Topic
- Software development
- Object-oriented Python
- Jupyter Notebooks
- Data ingestion, manipulation, and cleaning using DataFrames
- Data exploration and interpretation using visualization and statistical techniques
- Training and applying basic machine learning models
- Ability to apply and interpret appropriate metrics for classification and regression problems
Coordinator Dr. RJ Nowling
Add to Portfolio (opens a new window)
|
|