6 ECTS credits
180 u studietijd

Aanbieding 1 met studiegidsnummer 4023972FNR voor alle studenten in het 2e semester met een gespecialiseerd master niveau.

Semester
2e semester
Inschrijving onder examencontract
Niet mogelijk
Beoordelingsvoet
Beoordeling (0 tot 20)
2e zittijd mogelijk
Ja
Onderwijstaal
Nederlands
Onder samenwerkingsakkoord
Onder interuniversitair akkoord mbt. opleiding
Faculteit
Faculteit Ingenieurswetenschappen
Verantwoordelijke vakgroep
IR Academische eenheid
Externe partnerinstelling(en)
Universiteit Gent
Onderwijsteam
Jeroen Van Schependom
Decaan IR (titularis)
Onderdelen en contacturen
22 contacturen Hoorcollege
48 contacturen Werkvormen en Praktische Oef.
3 contacturen Zelfwerk en -studie
Inhoud

Zie Studiefiche E010620 UGent

 

Position of the course

This course teaches the principles of computational neuroscience and applies these principles in practica, which cover the implementation and study of neuronal models. The course starts from how numerical models have been developed to describe experimental neuroscience data and to study the spiking properties of single neurons within the brain. Afterwards, the dynamics of neurons within a population or network are discussed to cover higher-level properties such as learning, adaptation or plasticity. The exercises/practica are based on realistic datasets and problem sets and teach an important computational skillset to those with an interest in neuroscience or in developing neuronal/brain-inspired technologies.  
 
Contents  

1. Theoretical basis

 Hodgkin-Huxley neuronal models: successes and limitations

 From experiment to model: spiking statistics, refractoriness, PSTH, PLV

 Neuronal codes: neurons in a network (plasticity, receptive fields, rate/synchrony)

 Neuronal plasticity and learning: Hebbian’s rule, unsupervised learning  

2. Application Oriented:

 Programming numerical methods in Python/Matlab.

 Hands-on experience with classical neuronal (network) models.

Bijkomende info

Zie Studiefiche E010620 UGent

Keywords  Computational neuroscience, Neuronal modeling, Neural codes, Spiking models and networks 

Initial competences  Signals and Systems, Mathematics & Statistics, Python or Matlab programming skills.

Conditions for credit contract  Access to this course unit via a credit contract is determined after successful competences assessment  
 
Conditions for exam contract  This course unit cannot be taken via an exam contract  
 
Teaching methods  lectures, exercises, practica, self-reliant study activities  
 
Extra information on the teaching methods  The course covers four topic modules which will last 3-4 weeks each. Each topic will be introduced during a lecture, after which small exercises are given and a practicum exercise is introduced (at UGent). The practica (with local UGent or VUB support) will involve programming exercises and neuronal model simulations and can be completed in student pairs. A written report needs to be prepared for each of the four practica.   
 
Learning materials and price  Slides, (e-)book chapters, publicly available publications/code repositories.  
 
References  Neuronal Dynamics - from single neurons to networks and models of cognition (W. Gerstner, W.M. Kistler, R. Naud and L. Paninski), Cambridge Univ. Press. 2014

Leerresultaten

Final Competences

Zie Studiefiche E010620 UGent

1. Knowledge of how neuronal models can be adopted to simulate or understand experimental neuroscience.

2. Model and understand neuronal characteristics of single unit neuron models (adaptation, refractoriness, spiking probabilities, PSTH, PLV).

3. Understand the limitations and parameter choices of descriptive neuronal population models (synchrony, coding principles, inhibition/excitation, receptive fields)

4. Knowledge of how neural plasticity and learning can be modelled.

5. Python/Matlab programming skills to implement and evaluate neuronal models.

Beoordelingsinformatie

De beoordeling bestaat uit volgende opdrachtcategorieën:
Examen Andere bepaalt 100% van het eindcijfer

Binnen de categorie Examen Andere dient men volgende opdrachten af te werken:

  • Exam+reports met een wegingsfactor 1 en aldus 100% van het totale eindcijfer.

    Toelichting: Activities/Reports during the semester (and feedback discussion) count for 60 %  

    Final written/oral exam counts for 40%.

    Zie Studiefiche E010620 UGent

Aanvullende info mbt evaluatie

Zie Studiefiche E010620 UGent

Evaluation methods  end-of-term evaluation and continuous assessment via reports  
 
Examination methods in case of periodic evaluation during the first examination period  Written examination, open book examination, oral examination  
 
Examination methods in case of periodic evaluation during the second examination period  Written examination, open book examination, oral examination  
 
Examination methods in case of permanent evaluation  Assignment  
 
Possibilities of retake in case of permanent evaluation  Examination during the second examination period is possible in modified form  
 
Extra information on the examination methods  
During examination period: oral open-book exam, graded project reports, contribution to tasks.  
 
Calculation of the examination mark

Activities/Reports during the semester (and feedback discussion) count for 60 %  

Final written/oral exam counts for 40%.

Toegestane onvoldoende
Kijk in het aanvullend OER van je faculteit na of een toegestane onvoldoende mogelijk is voor dit opleidingsonderdeel.

Academische context

Deze aanbieding maakt deel uit van de volgende studieplannen:
Master in de ingenieurswetenschappen: biomedische ingenieurstechnieken: Standaard traject