Credited Courses

The Center for Advanced Research Computing (CARC) offers a 2-credit Technology and Applied Computing (TAC) course—TAC-450: High-Performance Computing in Applied Machine Learning—through the USC Viterbi School of Engineering. The class runs in the Fall semester only.

Through lectures, interactive hands-on sessions, assignments, and a team project, students learn how to apply cutting-edge tools and technologies in high-performance computing, machine learning, and deep learning to solve real-world science and engineering problems while working with modern computing systems.

Students work on one team project with three to four members, to complete by the end of the semester. For the team project, students use state-of-the-art computing resources at CARC to analyze real-world datasets using the techniques discussed in class, extract valuable insights, and present their descriptive or predictive models to the class.

The main objectives of this course are to:

• Explain the hardware and software components of an HPC cluster and how to use them effectively.
• Compare CPU and GPU architectures and identify which one would be a more suitable choice in a given scenario.
• Develop codes in Python, using CUDA-enabled libraries, to perform data analysis and machine learning tasks on GPUs.
• Explain fundamental concepts in deep and reinforcement learning and implement those in Python scripts, using appropriate libraries such as PyTorch and Keras, to solve real-world science and engineering problems.
• Demonstrate an understanding of cloud infrastructure (in AWS), identify the services required to perform data analysis and deep learning tasks, and utilize them to address practical science and engineering problems.

Prerequisites: TAC-449, DSCI-352, MATH-449, BUAD-425, or CSCI-467

Interested students should visit the USC Viterbi School of Engineering TAC page for further information.