Leren

Learning

Eigenaar	Bachelor Kunstmatige Intelligentie
Coördinator	dr. E.T. Nalisnick
Onderdeel van	Minor Kunstmatige Intelligentie, jaar 1Exchange Programme Faculty of Science, specialisation BSc Artificial Intelligence, jaar 1Bachelor Kunstmatige Intelligentie, jaar 2Bachelor Bèta-gamma, major Kunstmatige Intelligentie, jaar 3Bachelor Future Planet Studies, major Kunstmatige Intelligentie, jaar 3

Studiewijzer 2020/2021

Globale inhoud

Machine learning is programming computers to optimize a performance criterion using example data or past experience. We need learning in cases where we cannot directly write a computer program to solve a given problem, but need example data or experience. One case where learning is necessary is when human expertise does not exist, or when humans are unable to explain their expertise. Already, there are many successful applications of machine learning in various domains: There are commercially available systems for recognizing speech and handwriting. Retail companies analyze their past sales data to learn their customers’ behavior to improve customer relationship management. Financial institutions analyze past transactions to predict customers’ credit risks. Robots learn to optimize their behavior to complete a task using minimum resources. These are some of the applications that we will discuss in this lecture. We can only imagine what future applications can be realized using machine learning: Cars that can drive themselves under different road and weather conditions, phones that can translate in real time to and from a foreign language, autonomous robots that can navigate in a new environment, for example, on the surface of another planet. Machine learning is certainly an exciting field to be working in! (from "Introduction to Machine Learning, 3rd Edition, MIT Press, Ethem Alpaydin").

Studiemateriaal

Literatuur

• Ethem Alpaydin, 'Introduction to Machine Learning', MIT Press, 3rd edition

Overig

• Course notes and slides on Canvas

Leerdoelen

• The student can summarize the primary goal of machine learning, the general methods by which this is achieved and the resulting inherent limitations.
• The student can categorize a problem description or algorithm as belonging to a specific type of machine learning, e.g. supervised vs. unsupervised, classification vs. regression, etc.
• The student can describe the following models: Linear Regression, Naive Bayes, K-Means, EM, PCA, k-NN, Decision Trees, Random Forests, Logistic Regression and Neural Networks; and for each of these models:
• The student can identify the type of problems or data sets the model would be most suited to.
• The student can apply the model to either discrete or continuous data sets and interpret the results.
• The student can indicate what elements determine the complexity of the hypotheses that can be modeled and which hyper-parameters affect this complexity.
• The student can define the quantity that is optimized by the model, how the optimization is achieved, and the potential limitations of an obtained optimum.
• The student can explain the mathematical assumptions made about the distribution of the data and use these to derive the optimization criterion of the model.
• The student can evaluate the fit of a learned model using cross-validation, diagnose overfitting or underfitting and use these results to tune and improve the model.
• The student can analyze a complete machine learning pipeline, including problem statement, data collection, data processing, model application and validation, and recommend a viable approach that would improve the overall result.

Onderwijsvormen

• Werkcollege
• Hoorcollege
• Laptopcollege
• Zelfstudie

Verdeling leeractiviteiten

Aanwezigheid

Aanwezigheidseisen opleiding (OER-B):

• Voor practica en werkgroepbijeenkomsten met opdrachten geldt een aanwezigheidsplicht. De invulling van deze aanwezigheidsplicht kan per vak verschillen en staat aangegeven in de studiewijzer. Wanneer studenten niet voldoen aan deze aanwezigheidsplicht kan het onderdeel niet met een voldoende worden afgerond.
  

Toetsing

Onderdeel en weging	Details
Eindcijfer
1 (100%) Deeltoets 1

There will be weekly assignments (written and programming) during the course. The course will conclude with one mid-term exam, one final exam and one retake exam.

Minimum required grade for the assignments is 4.5, if the student fails to obtain a minimum of 4.5 from the assignments (written and programming together), the student will fail the course.
Minimum required grade for the exams is 4.5, if the student fails to obtain a minimum of 4.5 (mid-term and final together), the student will fail the course.

The grade will be computed as follows:

Total = 0.15*mid-term + 0.25*final + 0.3*written + 0.3*programming

If the student fails, the retake exam will effect the grade as follows:

Total = 0.6*retake + 0.2*written + 0.2*programming

Fraude en plagiaat

Dit vak hanteert de algemene 'Fraude- en plagiaatregeling' van de UvA. Hier wordt nauwkeurig op gecontroleerd. Bij verdenking van fraude of plagiaat wordt de examencommissie van de opleiding ingeschakeld. Zie de Fraude- en plagiaatregeling van de UvA: http://student.uva.nl

Weekplanning

Rooster

Het rooster van dit vak is in te zien op DataNose.

Aanvullende informatie

Recommended prior knowledge: having successfully completed a. BSc level Linear Algebra course as well as a BSc level Statistics course is a requirement. Without basic math and statistics knowledge this course can not be successfully completed. 

Course is taught in English

Verwerking vakevaluaties

Hieronder vind je de aanpassingen in de opzet van het vak naar aanleiding van de vakevaluaties.

Contactinformatie

Coördinator

• Tim Doolan

Coördinator

• dr. Eric Nalisnick

Docenten

• Mattijs Blankesteijn BSc
• Mohammad Derakhshani MSc
• R.T.A. van Drimmelen
• Micha de Groot BSc
• A.C. Groot BSc
• Samar Hashemi
• Andy Keller MSc
• Alex Khawalid
• David Knigge BSc
• Thijs Kuipers
• Putri van der Linden MSc
• Sabijn Perdijk
• Bas Terwijn
• Siem Teusink
• Li-anne Tjin