COMPX525-22A (HAM)

Deep Learning

15 Points

Edit Header Content
Division of Health Engineering Computing & Science
School of Computing and Mathematical Sciences
Department of Computer Science

Staff

Edit Staff Content

Convenor(s)

Bernhard Pfahringer

Bernhard Pfahringer
4041
G.2.23
See Moodle
bernhard.pfahringer@waikato.ac.nz

Lecturer(s)

Administrator(s)

: maria.admiraal@waikato.ac.nz
: buddhika.subasinghe@waikato.ac.nz

Placement/WIL Coordinator(s)

Tutor(s)

Student Representative(s)

Lab Technician(s)

Librarian(s)

: alistair.lamb@waikato.ac.nz

You can contact staff by:

  • Calling +64 7 838 4466 select option 1, then enter the extension.
  • Extensions starting with 4, 5, 9 or 3 can also be direct dialled:
    • For extensions starting with 4: dial +64 7 838 extension.
    • For extensions starting with 5: dial +64 7 858 extension.
    • For extensions starting with 9: dial +64 7 837 extension.
    • For extensions starting with 3: dial +64 7 2620 + the last 3 digits of the extension e.g. 3123 = +64 7 262 0123.
Edit Staff Content

Paper Description

Edit Paper Description Content

This paper provides an introduction into Deep Learning, focussing on both algorithms and applications.

The learning outcomes for this paper are linked to Washington Accord graduate attributes WA1-WA11. Explanation of the graduate attributes can be found at: https://www.ieagreements.org/

Edit Paper Description Content

Paper Structure

Edit Paper Structure Content

Viewing of all recorded material is expected, as is participation in the lectures/Q+A sessions. You are responsible for being familiar with all material covered.

Edit Paper Structure Content

Learning Outcomes

Edit Learning Outcomes Content

Students who successfully complete the paper should be able to:

  • Demonstrate good understanding of the basics of Deep Learning (Test) [WA1, WA7]
    Linked to the following assessments:
  • Demonstrate familiarity with the current mainstream Deep Learning algorithms (Test) [WA1, WA7]
    Linked to the following assessments:
  • Build a DL-based image classification from scratch (Assignment + Competition) [WA2, WA3, WA4, WA5, WA9]
    Linked to the following assessments:
  • Employ and adapt pretrained models in DL applications (Assignment + Competition) [WA2, WA3, WA4, WA5, WA9]
    Linked to the following assessments:
Edit Learning Outcomes Content
Edit Learning Outcomes Content

Assessment

Edit Assessments Content

The assessment is based on the four components specified below.

If you are enrolled in a BE(Hons), samples of your work may be required as part of the Engineering New Zealand accreditation process for BE(Hons) degrees. Any samples taken will have the student name and ID redacted. If you do not want samples of your work collected then please email the engineering administrator, Natalie Shaw (natalie.shaw@waikato.ac.nz), to opt out.
Edit Additional Assessment Information Content

Assessment Components

Edit Assessments Content

The internal assessment/exam ratio (as stated in the University Calendar) is 100:0. There is no final exam. The final exam makes up 0% of the overall mark.

The internal assessment/exam ratio (as stated in the University Calendar) is 100:0 or 0:0, whichever is more favourable for the student. The final exam makes up either 0% or 0% of the overall mark.

Component DescriptionDue Date TimePercentage of overall markSubmission MethodCompulsory
1. 12 short Moodle quizzes (one every lecturing week)
10 Jun 2022
No set time
24
  • Online: Submit through Moodle
2. Programming assignment
10 Jun 2022
No set time
25
  • Online: Submit through Moodle
3. Kaggle competition
27 May 2022
No set time
25
  • Online: Submit through Moodle
4. Test
14 Jun 2022
4:00 PM
26
  • Online: Submit through Moodle
Assessment Total:     100    
Failing to complete a compulsory assessment component of a paper will result in an IC grade
Edit Assessments Content

Required and Recommended Readings

Edit Required Readings Content

Required Readings

Edit Required Readings Content
There is no required text for this paper.
Edit Required Readings Content

Recommended Readings

Edit Recommended Readings Content

There are way too many books on deep learning. Here are my current three favourites:

https://www.deeplearningbook.org

https://d2l.ai

https://github.com/fastai/fastbook

Edit Recommended Readings Content

Other Resources

Edit Other Resources Content
Supporting material will be provided through Moodle, as appropriate.
Edit Other Resources Content

Online Support

Edit Online Support Content
via Moodle
Edit Online Support Content

Workload

Edit Workload Content
The expected workload for this paper is around 10 hours for each week during Trimester A, which comprises lecture participation, viewing of recorded material, background reading, and working on the programming assignment and competition.
Edit Workload Content

Linkages to Other Papers

Edit Linkages Content
This paper builds on prior knowledge coming from both COMPX305 and COMPX310. There may be some overlap with COMPX521, but it will definitely be complimentary to any other Machine Learning related Level 5 offering.
Edit Linkages Content

Prerequisite(s)

Prerequisite papers: COMPX310

Corequisite(s)

Equivalent(s)

Restriction(s)

Edit Linkages Content