IN-STK5000 – Responsible Data Science
Course description
Schedule, syllabus and examination date
Course content
Classic approaches in data analysis are based on a static (or predefined) procedure for both collecting and processing data. Modern approaches deal with the adaptive procedures which in practice almost always are used.
In this course you will learn how to design systems that adaptively collect and process data in order to make decisions autonomously or in collaboration with humans. The course applies core principles from machine learning, artificial intelligence, databases and parallel computing to real-world problems in safety, reproducibility, transparency, privacy and fairness.
Learning outcome
After taking the course, you will:
- See adaptive data analysis holistically, as a general decision problem.
- Know how to adaptively plan data collection.
- Understand when privacy is an issue and how to deal with privacy concerns.
- Provide transparency by quantifying the strength of conclusions and ensuring reproducibility.
- Be able to provide safety and reliability guarantees.
- Have insight into issues of discrimination and fairness that can arise.
- Be able to use large-scale data processing tools such as Tensor-Flow
- Be able to deal with outliers, data contamination, etc.
Admission to the course
IN-STK5000 and IN-STK9000 are viewed together in relation to admission and available spots. If the number of enrolled students is higher than the number of available spots, they will be ranked as follows:
- PhD candidates who have the topic approved in their study plan
- Master?s students at the master?s program Computer Science who have passed the course approved in their curriculum
- Master?s students at the Faculty of Mathematics and Natural Sciences who have approved the subject in their curriculum
- Master?s students at the Faculty of Mathematics and Natural Sciences
- Other
Recommended previous knowledge
Knowledge in probability (STK1000 – Introduction to Applied Statistics or STK1100 – Probability and Statistical Modelling) or Discrete Mathematics. Elementary calculus (differentiation, integration). Elementary programming skills (Python)
Some basic mathematical knowledge in probability (e.g.?STK1100),?linear algebra (e.g.?MAT1120 – Linear Algebra) and algoritms (e.g.?IN2010 – Algorithms and Data Structures)
Overlapping courses
- 10 credits overlap with IN-STK9000 – Adaptive methods for data-based decision making.
Teaching
4 hours of lectures/exercises/lab each week for the whole semester.
More about mandatory assignments and other handouts.
Examination
Submission of mandatory assignments is required. All?mandatory assignments prior to an exam must be approved?before you can take the exam.
Exam format: Two group reports, based on two mini-projects, and a final oral/written exam.
Each report constitutes 35% of the final grade and the final exam constitutes 30% of the final grade. All parts must have a pass grade, and all parts must be passed in the same semester.
Examination support material
All written material allowed
Language of examination
The examination text is given in English, and you submit your response in English.
Grading scale
Grades are awarded on a scale from A to F, where A is the best grade and F is a fail. Read more about the grading system.
Resit an examination
Students who can document a valid reason for absence from the regular examination are offered a postponed examination. Re-scheduled examinations are not offered to students who withdraw during, or did not pass the original examination.
More about examinations at UiO
- Use of sources and citations
- Special exam arrangements due to individual needs
- Withdrawal from an exam
- Illness at exams / postponed exams
- Explanation of grades and appeals
- Resitting an exam
- Cheating/attempted cheating
You will find further guides and resources at the web page on examinations at UiO.