DDCS: Data Driven Computer Science - COMS20011 23-24

The course will teach the basics of how to reason with data!

Staff

• Majid Mirmehdi [m.mirmehdi@bristol.ac.uk] (unit director)
• Charles Kind [charles.kind@bristol.ac.uk]

Schedule:

• Lectures:
  • Thursdays 2pm in QUEENS BLDG 1.40 PUGSLEY
  • Fridays 10am in QUEENS BLDG 1.40 PUGSLEY
• Labs (starting Jan 29th). To tell what group you're in look in your Outlook calendar. You will see one of the following two slots:
  • Group 1: Thursday 1-2pm in MVB 2.11
  • Group 2: Thursday 2-3pm in MVB 2.11

Lecture recordings:

All lecture recordings eventually appear on the Re/Play system in BlackBoard afew hours after the event. Click on Re/Play on the left menu bar on the BB course webpage to find them.

Labs:

• The labs are starting on Jan 29th, so there will be no lab on Jan 22nd.
• The labs come with exercises as Python notebook. A new set of exercises will be released every TWO weeks.
• Material in the labs may appear in the exam.
• The labs are the primary route for feedback on all aspects of the course, including lectures, any question sheets and exams, not just the labs!!!!
• The lab exercises will be released in the /labs/ folder. @@ Lab1 is now available @@
• TAs: Xuan kong, Zheng Dong, Lucas Mockeridge, Harleen Gulati, Austin Long, Deep Pawar, Jeein Park

Teams:

• The Teams Team is [COMS20011: Data-Driven Computer Science 2023/24]. You should have access already through the "Teams" panel. If not, please get in touch with the COMS Student Enquiries Mailbox at coms-student-enquiries@bristol.ac.uk

Exam info:

• The course is assessed 100% by a multiple-choice exam.
• The exam will have 20 questions, worth 100 marks, and will be 2 hours. It does require a calculator (which must be non-programmable).
• If you have any questions regarding alternative exam arrangements, please get in touch with the COMS Student Enquiries Mailbox at coms-student-enquiries@bristol.ac.uk
• The exam is closed-book (so no additional materials are allowed). There will be a formulae sheet with the exam.

Past exam papers:

• Example past papers will be released soon after Week 20.

---

Weekly Lecture materials

Introduction (Majid)

Week 13: 22/01/2024

Lecture, etc.	Slides, etc.
MM01. Intro to COMS20011 - a bit fishy!	[pdf]
MM02. Sampling, Distance Measures	[pdf]
Problem Sheet 01 (Self/Group study)	[pdf]
Problem Sheet 01 (Solutions)	[pdf]

Week 14: 29/01/2024

Lecture, etc.	Slides, etc.
MM03. Data Characteristics, Covariance, Eigen Analysis, Outliers	[pdf]
Problem Sheet 02 (Self/Group study)	[pdf]
Problem Sheet 02 (Solutions)	[pdf]
CK01. Mathematical prerequisites, Calculus	[pdf]

Week 15: 05/02/2024

Lecture, etc.	Slides, etc.
CK02. Mathematical prerequisites, Series, Logs and exponentials	[pdf]
CK03. Mathematical prerequisites, Vectors, Matrices and Kronecker delta	[pdf]

Week 16: 12/02/2024

Lecture, etc.	Slides, etc.
CK04. Problems from mathematical prerequisites. Regression, loss and curve fitting	[pdf]
CK05. Regression, loss and curve fitting, one dimensional	[pdf]

Week 17: 19/02/2024

Lecture, etc.	Slides, etc.
CK06. Regression, loss and curve fitting, many dimensions	[pdf]
CK07. Maximum likelihood	[pdf]
Probabilistic Machine Learning	[pdf]
Mathematics for Machine Learning	[pdf]

Week 18: 26/02/2024

Reading Week

Week 19: 04/03/2024

Lecture, etc.	Slides, etc.
CK08 and CK09. Maximum likelihood	[pdf]

Week 20: 11/03/2024

Lecture, etc.	Slides, etc.
MM04. Signals & Frequencies	[pdf]

Week 21: 18/03/2024

Lecture, etc.	Slides, etc.
Problem Sheet 03 (Self/Group study)	[[pdf]](released on Thursday)

Week 22: 15/04/2024

Week 23: 22/04/2024

Week 24: 29/04/2024