Dipartimento di InformaticaUniversità degli Studi di Milano
Dario Malchiodi

Architectural patterns for distributed machine learning applications2019/20

PhD in Computer Science · University of Milan · Lectures are in English · 4 hours

Archived edition: this page is kept for reference, and some materials or links may no longer be available.

Professors: Nicolò Cesa-Bianchi (unimi), Andrea Condorelli (Magneti Marelli), Dario Malchiodi (unimi).

News

  1. Cancellation of teaching activities

    All teaching activities are canceled until 29/2.

  2. Schedule change for the Architecturel patterns for distributed machine learning application course

    The lecture of the course Architectural patterns for distributed machine learning application of 26/2 is postponed on 2/3, between 15:30 and 18:30 in the laboratorio magistrale classroom of the third floor.

  3. Classroom change for the Architecturel patterns for distributed machine learning application course

    The lectures of the course Architectural patterns for distributed machine learning application will take place in different classrooms than previously announced. Students are invited to check the new course timetable.

Course schedule

Lectures take place in the meeting room in the Computer Science department, according to the following schedule:

Day Hour Place
10/02/2020 14:30 - 16:30 sala riunioni 3° piano
11/02/2020 14:30 - 16:30 sala riunioni 5° piano
13/02/2020 14:30 - 16:30 sala riunioni 7° piano
14/02/2020 14:30 - 16:30 sala riunioni 7° piano
17/02/2020 15:30 - 18:30 lab. laurea magistrale 5° piano
21/02/2020 15:30 - 18:30 lab. laurea magistrale 3° piano
24/02/2020 15:30 - 18:30 lab. laurea magistrale 5° piano
26/02/2020 02/03/2020 15:30 - 18:30 lab. laurea magistrale 5° piano 3° piano

Any change to the schedule will be announced in class and published in paragraph News of this page. There will be additional, non-compulsory, lab lectures each Friday, starting mid-october. The starting date will be communicated later on.

Office hours

By appointment, room 5015 of the Computer Science Department. It is possible contact the teacher by e-mail, taking care to read in advance the guide prepared by Prof. Sebastiano Vigna and clearly specifying in the message the course name and the academic year. In particular, students are encouraged to always use their academic address (i.e. based on the domain studenti.unimi.it) signing with name and student ID number and recalling that the response time may vary depending on the teacher commitments.

Course material

  • The part on federated learning is described in the presentation used during the first two lectures.
  • The part on architectures is described in the presentations used during the fifth and sixth lecture, and in the following papers:
    • J. A. Zachman, A framework for information systems architecture, IBM Systems Journal 26-3 (1987), 454-470;
    • D. E. Perry, A. L. Wolf, Foundations for the Study of Software Architecture, ACM SIGSOFT 17-4 (1992), 40-52;
    • M. Shaw, Larger Scale Systems Require Higher-Level Abstractions, ACM SIGSOFT 14-3 (1989), 143-146.
    • S. Ghemawat, H Gobioff, S. Leung, The Google file system, Proceedings of the nineteenth ACM symposium on Operating systems principles, 2003, 29-43.
    • K. Aida, Effect of job size characteristics on job scheduling performance, in Workshop on Job Scheduling Strategies for Parallel Processing, Springer, Berlin, Heidelberg, 2000, 1-17.
    • A. Sharma, M. Kumar, S. Agarwal, A complete survey on software architectural styles and patterns, Procedia Computer Science 70, 2015, 16-28.
    • J. Dean, S. Ghemawat, MapReduce: simplified data processing on large clusters. Communications of the ACM 51.1, 2008, 107-113.