Machine Learning - Based Automated Performance Tuning2014HT
|
|
Course plan
Organization
Lectures (ca. 12h), student projects and/or presentations. Written exam.
Recommended for
Graduate (CUGS, CIS, ...) students interested in the application of machine learning techniques to advanced system performance optimization, as in compiler construction, library generation, runtime systems, parallel computing, software engineering, system simulation and optimization.
The course was last given
HT 2012.
Goals
The course introduces fundamental techniques of machine learning and considers case studies for its application in automated system performance tuning, such as auto-tuning library generators, compilers, and runtime systems.
Prerequisites
Linear algebra. Discrete mathematics. Data structures and algorithms. Some basic knowledge of computer architecture is assumed. For the case study presentations, some background in at least one application area, such as compiler construction, library generation, signal processing software, runtime systems, or software composition, is required.
Contents
"[Machine] learning is the process of [automatically] constructing, from
training data, a fast and/or compact surrogate function that heuristically
solves a decision, prediction or classification problem for which only
expensive or no algorithmic solutions are known. It automatically abstracts
from sample data to a total decision function."
- [Danylenko et al., Comparing Machine Learning Approaches..., SC'2011, LNCS
6708]
The course will introduce basic machine learning techniques together with
principles of autotuning, and emphasize on the application of machine learning
in performance autotuning (student projects and/or presentations).
Organization
Lecture block(s) (several days) and presentation session (1 day) in Linköping.
Literature
To be announced.
Lecturers
Christoph Kessler, Welf Löwe.
Examiner
Christoph Kessler.
Examination
Written exam, 1.5p
Small project with presentation or presentation of a research paper, 1.5p
Credit
3hp if both examination moments are fulfilled. Admission to the exam requires attendance in 50% of the lectures and lessons.
Organized by
CUGS
Comments
New course 2012.
Page responsible: Director of Graduate Studies