Neural Computation 2010-2011

Neural Computation (NC) is a 10 point MSc course of 18 lectures in the first semester.
Lectures are Tuesday and Friday: 12:10-13:00.
Instructor: James A. Bednar

Short description

In this course we study the computations carried out by the nervous system. Unlike other courses and typical implementations of artifical neural networks, we take a bottom-up approach. This means that we incorporate data from neurobiology, simulate certain aspects of it, and try to formulate theories about the brain.

Keywords: single neuron models, neural codes, plasticity models.

Copies of the lecture notes part1, part2 will be handed out in the lectures. Many of the cited references can be found using PubMed. Pages from previous years are available for reference.

Office hours: make an appointment or catch me after the lecture.

Prerequisites

No prior biology/neuroscience knowledge is required. The tutorials and assignments use a small subset of basic University-level math. These older FMCS lecture notes can be used as a refresher. Alternatively, use Google to look up forgotten maths if needed. If you are still stuck, use the practicals or office hours to resolve the problems.

In the tutorials we use MatLab and NEURON (a special purpose simulator). No prior experience with either is required, but note that MatLab skills are valuable for many courses. We provide some notes on using Matlab.

Assessment

The course will be fully assessed by two reports of practical assignments that will appear here (deadlines will be announced). The two marks are averaged to get the final course mark. Standard late policies will apply, namely that late coursework is not accepted without good reason, which must be discussed with the MSc Course Organiser, not the lecturer. Also see How to make graphs and write reports and Formal Writing Tips.

Assignment 1: Spike propagation. Deadline 11 Nov 2010 at 4pm.

Assignment 2: BCM plasticity. Deadline 13 Jan 2011 at 4pm.

Practicals

Practicals are scheduled for Wednesdays 16:00-18:00 or Fridays 14:00-16:00 (choose either one) in AT-4.14. There are no practicals in the first week. Tutors are Rui Costa and Helen Ramsden.

Timetable (approximate)

Week 1; week of Sep 20
Tuesday lecture: 1. Introduction and Chapter 1: Anatomy
Friday lecture: NO LECTURE
No practical.

Week 2; week of Sep 27
Tuesday lecture: 2. Chapter 2: Passive properties.
Friday lecture: 3. Chapter 3: Hodgkin-Huxley
Practical: 1. The NEURON simulator: Passive properties

Week 3; week of Oct 4
Tuesday lecture: 4. Chapter 3: Hodgkin-Huxley
Friday lecture: 5. Chapter 4: Synapses
No practical.

Week 4; week of Oct 11
Tuesday lecture: 6. Chapter 4: Synapses
Friday lecture: 7. Chapter 5: Integrate and Fire
Practical: 2. The NEURON simulator: Hodgkin-Huxley model

Week 5; week of Oct 18
Tuesday lecture: 8. Chapter 6: Firing statistics
Friday lecture: 9. Chapter 7: Retina and V1
Practical: 3. NEURON: Interactions of synapses on dendrites

Week 6; week of Oct 25
Tuesday lecture: 10. Chapter 7: Retina and V1
Friday lecture: 11. Chapter 8: Higher visual processing
Practical: 4. Matlab: AMPA receptor simulation. Script: ampa.m

Week 7; week of Nov 1
Tuesday lecture: 12. Chapter 9: Population codes
Friday lecture: 13. Chapter 9: Coding
Practical: 5. Matlab: Mutual information of a Poisson spiker. Script: bits.m

Week 8; week of Nov 8
Tuesday lecture: 14. Chapter 10: Networks (see Dayan and Abbott, Theoretical Neuroscience, Chapter 7)
Friday lecture: NO LECTURE -- At SfN
Practical: 6. Matlab: Ben-Yishai network. Script: ben2.m

Week 9; week of Nov 15
Tuesday lecture: NO LECTURE -- At SfN
Friday lecture: NO LECTURE -- At SfN

Week 10; week of Nov 22
Tuesday lecture: 15. Chapter 11+12: Spiking networks, Decisions
Friday lecture: 16. Chapter 13: Hebbian learning
No practical.

Week 11; week of Nov 29
Tuesday lecture: 17. Chapter 14: Spike timing dependent plasticity
Friday lecture: 18. Semester overview
Practical: 7. Matlab: Hebbian learning with constraints. Script: hebb.m

Additional information:

Additional Matlab: An Integrate and fire neuron.  Script: mvr_if_matlab.m

Movies of LGN and V1 recordings (play with mplayer under linux):

hubel_Wiesel_lgn_off_cell.asf
hubel_wiesel_binocular_cell.asf
hubel_wiesel_complex.asf
hubel_wiesel_directional_cell.asf
hubel_wiesel_lgn_on_cell.asf
hubel_wiesel_simple_cell.asf

Recurrent 6-node network with chaotic behavior bifur6.m


Home : Teaching : Courses : Nc 

Informatics Forum, 10 Crichton Street, Edinburgh, EH8 9AB, Scotland, UK
Tel: +44 131 651 5661, Fax: +44 131 651 1426, E-mail: school-office@inf.ed.ac.uk
Please contact our webadmin with any comments or corrections. Logging and Cookies
Unless explicitly stated otherwise, all material is copyright © The University of Edinburgh