SluitenHelpPrint
Switch to English
Cursus: INFOMLHVL
INFOMLHVL
Machine learning for human vision and language
Cursus informatieRooster
CursuscodeINFOMLHVL
Studiepunten (ECTS)7,5
Categorie / NiveauM (Master)
CursustypeCursorisch onderwijs
VoertaalEngels
Aangeboden doorFaculteit Betawetenschappen; Graduate School of Natural Sciences; Graduate School of Natural Sciences;
Contactpersoondr. B.M. Harvey
E-mailB.M.Harvey@uu.nl
Docenten
Docent
dr. T. Deoskar
Overige cursussen docent
Docent
dr. B.M. Harvey
Overige cursussen docent
Blok
1-GS  (06-09-2021 t/m 12-11-2021)
Aanvangsblok
1-GS
TimeslotD: D (WO-middag, WO-namiddag, Vrijdag)
Onderwijsvorm
Voltijd
Cursusinschrijving geopendvanaf 31-05-2021 t/m 27-06-2021
AanmeldingsprocedureOsiris
Inschrijven via OSIRISJa
Inschrijven voor bijvakkersJa
VoorinschrijvingNee
Na-inschrijvingJa
Na-inschrijving geopendvanaf 23-08-2021 t/m 20-09-2021
WachtlijstJa
Plaatsingsprocedureadministratie onderwijsinstituut
Cursusdoelen

At the end of the course, the student will be able to:

  • Explain the broad concepts behind deep learning from both computer science and neuroscience perspectives.
  • Explain deep learning’s advantages and limitations compared to other modelling and machine learning approaches.
  • Identify problems that deep learning is suited to addressing in the fields of cognitive (neuro-) science, linguistics and artificial intelligence
  • Design and implement deep learning approaches to address some problems in the domain of image and language processing.


You will be able to better understand literature on deep learning and its applications to cognitive science.
You will be in a good position to start gaining hands-on experience in a supervised or team setting, such as an internship or Master’s thesis project.

Assessment
The course goals will be examined in the following ways:

  • Students will attend lectures introducing the approach taken in deep learning systems, comparing this to how deep learning is implemented in biological brains, and introducing the main applications of deep learning to cognitive science and linguistics. Their understanding of this content will be assessed in a final exam.
  • Students will participate in discussions and reviews of relevant literature, which will be graded.
  • Students will work through lab practical assignments on visual processing and on language processing. The resulting reports will be graded.

To take part in the make up test the grade of the original test needs to be a 4 at least.


 

Inhoud
Machine learning with deep convolutional neural networks (deep learning) is being applied increasingly broadly in computer science, technology and scientific research.
This method allows computer systems to perform tasks that have previously been impossible or inaccurate for computers, but typically straightforward for humans.
Tasks like visual object identification and natural language processing have traditionally been investigated by cognitive scientists and linguists, but recent applications of deep learning to these tasks also positions them at the center of recent artificial intelligence developments.
Therefore, it is important for AI students and researchers to understand the links between cognitive science and AI.

In this course, you will learn the principles behind deep learning, an approach inspired by the structure of the brain.
You will learn how these principles are implemented in the brain, focusing on the two aspects of visual processing and language (semantic or syntactic) processing.
You will build your own deep learning systems for the interpretation of natural images and language, using modern high-level neural network APIs that make implementation of these systems accessible and efficient.

Course form
Lectures and tutorials.


 
Competenties
-
Ingangseisen
-
Voorkennis
This course is open to all students in the AI Master’s program.
For students from other programs it is advised that they have some interest and familiarity with (cognitive) psychology, (formal) linguistics, and programming.
When in doubt, please contact the course coordinator.
Verplicht materiaal
Artikelen
Selected papers
Werkvormen
Hoorcollege

Werkcollege

Toetsen
Eindresultaat
Weging100
Minimum cijfer-

SluitenHelpPrint
Switch to English