CS3332 Applied Machine Learning

Survey of machine-learning techniques of artificial intelligence with a particular focus on military applications.  Topics include types of machine learning, training and testing of machine learning, data preparation, decision trees, Bayesian reasoning, linear models, neural networks, case-based reasoning, and reinforcement learning.  Each method will be related to important military and government applications.  This course is intended for students who are not computer-science majors.

Prerequisite

CS3331

Lecture Hours

4

Lab Hours

0

Course Learning Outcomes

  • Define and recognize key machine-learning techniques and be able to explain them, including:
    • Caching, case-based reasoning, and decision trees
    • Concept learning of logical expressions
    • Classification using probabilistic reasoning
    • Neural networks, including convolutional and transformer-based
    • Generative adversarial learning
    • Heuristic search
    • Evolutionary computing
    • Ensemble learning
  • Recommend appropriate machine-learning techniques for key applications including advisory systems, planning systems, natural-language understanding, computer vision, and sensor systems.
  • Be able to clean and transform raw data to get it into a form efficiently usable by machine learning.
  • Be able to apply software packages for key machine-learning techniques to real data and analyze their results.
  • Be able to implement learning methods using a software tool.
  • Identify the major difficulties in implementing and testing learning systems, including explaining reasoning, bias, and adversarial attacks.