Treffer: Parallel value iteration as an introduction to undergraduate-level parallel computing courses.

Value iteration is often used as an introduction topic to reinforcement learning and optimal control. Value iteration can be used to solve both Markovian and deterministic processes. Here, we use value iteration to solve a specific deterministic optimal control problem, i.e., finding shortest paths from anywhere in a map to a specific target in that map. Considering the size of the selected maps, parallelizing value iteration may become inevitable. Thus, we explore five different parallelization frameworks that is available in Python programming language (i.e., vectorization, multiprocessing, Cython with OpenMP, Numba, and Open MPI) since Python programming language is known to be slow in executing codes with loop and even slower with nested loops. Additionally, solving one problem but with several different methods can be beneficial for pedagogical purposes. In this paper, we present the results of our explorations and use them to promote case-oriented methods (project-based learning) in undergraduate parallel computing courses. [ABSTRACT FROM AUTHOR]

Copyright of AIP Conference Proceedings is the property of American Institute of Physics and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)