Untitled Page

Synaptic plasticity is the change in synaptic connections between two neurons due to the activity of one or both of these neurons. It is believed to be the basis of learning memory and some forms of brain development. The course will study both abstract models and biophysical models of synaptic plasticity. Abstract models of synaptic plasticity demonstrate how the concept of synaptic plasticity can contribute to different forms of learning, memory and development and how this might contribute to machine learning. Biophysical models of synaptic plasticity are based on actual cellular and molecular mechanisms observed in neurons and demonstrate how synaptic plasticity can arise from real biological mechanisms. The class will also have guest lectures from experimentalists working in this field.

Estimated Course Outline:

1. Introduction.

Class 1 1/23/2006

The Synaptic hypothesis in learning, memory and development
Experimental evidence for the synaptic hypothesis
Synaptic plasticity
Examples of formal simplified models
What are the basic unresolved questions of the synaptic basis of learning and memory?

Lecture 1.ppt

2. Formal Models of learning and memory

Unsupervised learning and receptive field development – Hebb, PCA, learning paradigms and can they account for receptive field development in visual cortex.

Lecture 2

In addition, refresh your memory of linear algebra and in particular of eigen-value equations.

Additional reading : Theory of Cortical plasticity

Chapter 1

Chapter 5

Class 3 – 2/6 Unsupervised learning – BCM and ICA

Lecture 3 (BCM)

Class 4 – 2/13

Objective function formulation + ICA

Chapter 3 (objective function)

Hyvarinen Oja review paper

Plastic Networks

Class 5 – 2/20

Suprevised learning (ppt)

Associative memory

Reinforcment learning

3. The Biophysics of synaptic plasticity

Class 6- 2/27

Guest lecture – Mike Mauk : Learning in the cerebellum.

Induction Protocols for LTP and LTD – rate based and spike timing based protocols for inducing bidirectional synaptic plasticity.
What is altered during induction – presynaptic change of release, Postsynaptic changes of conductance and the silent synapse hypothesis.
Statistical methods for detecting the site of induction.
The Biophysical basis of induction – Calcium influx, NMDA receptors, Phosphatases and Kinases.

Synaptic plasticity intro

* 3/13 – Spring Break in UT Austin

Class 8- 3/20

Guest Lecture – Dan Johnston: Back propagating action potentials and LTP

.Continueon Synaptic Plasticity Intro.

Class 9- 3/27 Mid term exam

Powerpoint slides

Mechanisms of Synaptic plasticity

Linear superposition

papers

Song et. al. 2000

Kempter et. al. 2000

Class 11 - 4/10

Biochemical modeling
Michaelis Menten kinetics
Calcium dependent learning models – Simplified calcium based models for the induction of synaptic plasticity and how they can account for induction protocols and receptive field plasticity.

Biochemical modeling

Calcium dependent synaptic plasticity

Class 12 – 4/17

Homeostasis and RF plasticity

4. Synaptic stability

Molecular Bi Stability

Clusters of interacting receptors can stabilize synaptic efficacies. Shouval HZ. Proc Natl Acad Sci U S A. 2005, early version + appendix and Movies

Class 13 - 4/24 - effect of ongoing activity on stability ofd memory

Discrete bounded weights

Fusi et. al. 2005

Class 13 – 4/24

The instability of memory due to ongoing activity. The Fusi-Abbott approach.

Class 14 - 5/1 Final Exam !

Grading

The grade will be based on 50% Homework, 20% midterm and 30% for a final exam or project. Class participation will be weighed in as an extra bonus.

Homework

There will be 6 HW problems. Grade will be calculated on the basis of the best 5.

HW can be completed after the due date. This can result in an additional 2/3 of the remaining grade. For example if somone got 50% on the first submission, he can resubmit and get at most an additional 66% of 50. So the maximal total in this case will be 50+33=83%

Houston A: 85-100 B: 70-85

Austin

Graduate:

A+: 95-100 A : 90-95 A -: 85-90 B+: 80-85 B : 75-80 B- : 70-75 C: 65-75 Below 60, Fail

Undergrad (377T)

Undergraduates taking the course as BME 377T will get a bonus 10% to all their grades (eg 70% => 77%) for the calculation of the final grade.

A: 90-100 B: 75-90 C: 65-75 Below 60, fail

Prerequisites:

This course will use mathematical methods such as linear algebra, Calculus, and Differential Equations. Therefore one semester of college level Calculus and Linear algebra are required. Many of the homework problems will be Matlab based. Therefore, some experience in programming is necessary. In addition basic knowledge in Neuroscience is required. Please consult with me if you are unsure that your knowledge in any of these areas is sufficient.

Time and Place