Abstract
The digitization and automation of business processes has increase the amount of digital data many folds over the past decade. Facebook alone ingest 500 terabytes of data every day into its storage system, generated through activities like photo upload, Likes, status updates etc. The physical limitations of a stand-alone system renders it impractical to be employed in solving large scale data analytic and machine learning problem. Scaling large scale data analytic and machine learning algorithms to multiple machine is inevitable in-order to make use of huge amount of data.
This praktikum will provide opportunity to solve one of the large scale distributed machine learning problem using Message Passing Interface (MPI) programming model. MPI is the de-facto standard for parallel programming of distributed memory system. The task will be to implement a method proposed in a recent research paper, validating the proposed results and proposing improvements or generating new ideas.
The project allows students to gain practical knowledge and capabilities in the respective topic area.- Students are encouraged to make a group of 2-3.
- Software implementation should be done in Python, Java or C++
- Employing Message Passing Interface to parallelize and scale across multiple machines
- Groups can start immediately.
- Each group is supposed to give at least two presentations:
- a first presentation about ongoing work, showing a first implementation and commenting on problems (around mid term),
- a final presentation of the whole work (end of term).
- Weekly update/discussion on the progress of selected project
Start: We start off on Monday, 20.04.2015.We will have a tutorial on parallel programming using Message Passing Interface. We will make groups and issue topics to be worked on.
Registration:
- You can register for topics from now on via email.
- For topics please contact Mohsan jameel
(Further) obligatory course dates:
- Monday, 20.04.2015 (2pm): Kick-Off
- Monday 10.05.2015(2pm): First-Ideas
Preliminaries: This course is especially suitable for students who attended one of the following lectures last term: Business Analytics, Machine Learning, Artificial Intelligence
Betreuer: Mohsan Jameel
