- Course description, goals, integrity, knowledge base
- Course number: 263-2300, 6 credits
- Spring 2015, lectures: M 10:15-12:00, CHN C14; Th 9:15-10:00 CAB G51; occasional substitute lectures: W 13:15-15:00 HG D3.2
- Instructor: Markus Püschel
- TAs:
- Alen Stojanov
- Daniele Spampinato
- Gagandeep Singh
- Only for project supervision: Georg Ofenbeck
Grading
- 40% research project
- 25% midterm
- 35% homework
- There is no final Exam
Research Project
| # |
Title |
Supervisor/s |
| 1 |
Smoothed Particle Hydrodynamics |
AS, DS |
| 2 |
Particle Filtering |
MP |
| 3 |
Simulating Dynamical Features of Escape Panic |
AS, DS |
| 4 |
Social Force Model |
AS, DS |
| 5 |
Numerical treatment of SODEs |
AS, DS |
| 6 |
Bayesian Belief Propagation (MP) |
MP |
| 7 |
Corner Detection using the Harris Algorithm |
MP |
| 8 |
Text recognition in natural images |
GO, GS |
| 9 |
Simulation of Semiconductor Nanostructure |
MP |
| 10 |
Deep Matching and DeepFlow |
MP |
| 11 |
AdaBoost |
MP |
| 12 |
SGM Stereo Matching |
GO, GS |
| 13 |
Splitting methods in Control |
GO, GS |
| 14 |
Stochastic Combinatorial Optimization |
AS, DS |
Midterm
15. April, 13:15 - 15:00, HG E3 (without solution, with solution).
Homework
Lectures (including pdfs)
| Date |
Slides |
Notes |
Other |
| 16.02 |
Course motivation, overview, organization |
|
|
| 19.02 |
Reminder basic concepts, cost/performance analysis link |
|
|
| 23.02 |
Architecture/Microarchitecture, operational intensity, Core 2/Core i7 |
|
Intel Optimization Manual |
| 26.02 |
Optimization for instruction level parallelism (ILP), Benchmarking |
|
|
| 02.03 |
Small benchmarking guide, compiler limitations |
|
|
| 05.03 |
Locality and caches, blocking MMM |
notes |
|
| 09.03 |
Roofline model |
notes |
paper |
| 12.03 |
Linear algebra, LAPACK, BLAS, fast MMM |
notes |
|
| 16.03 |
Fast MMM continued |
|
|
| 19.03 |
Fast MMM continued |
|
|
| 23.03 |
Fast MMM: register renaming and virtual memory |
notes |
|
| 26.03 |
Memory bound computations, sparse MVM |
|
|
| 30.03 |
Sparse MVM continued, SIMD vectorization |
|
|
| 02.04 |
SIMD vectorization |
|
Intel Intrinsics Guide |
| 06.04 |
Spring break |
|
|
| 08.04 |
Spring break |
|
|
| 13.04 |
(Sechseläuten) class cancelled |
|
|
| 15.04 |
Midterm exam, HG E3 |
|
|
| 16.04 |
SIMD vectorization |
|
|
| 20.04 |
Compiler vectorization, linear transforms, fast Fourier transform (FFT) |
notes |
ICC vectorizations |
| 23.04 |
FFT |
notes |
|
| 27.04 |
Fast FFT implementation, FFTW |
notes |
|
| 30.04 |
cancelled (due to one-on-ones a week later) |
|
|
| 04.05 |
Spiral: DSL-based program generator for linear transforms |
|
|
| 07.05 |
cancelled (one-on-ones) |
|
|
| 11.05 |
Machine learning in autotuning |
|
|
| 14.05 |
(Ascension day) no class |
|
|
| 18.05 |
Language environments for building program generators |
|
|
| 21.05 |
no class (one-on-ones) |
|
|
| 25.05 |
(Whitsuntide) no class |
|
|
| 27.05 |
Project presentations (HG D3.2) |
|
|
| 28.05 |
Project presentations (CAB G11), 9:15-11:00 (one hour longer) |
|
|
