^CSE454^
[plan]
>2003>
CSE454/Local/2002/
Where and when: R2, Semester 1, weeks 1-7, Tuesday, 3-5pm.
- week 1, Tuesday 5 March
[admin]
[intro]
[data]
[information]
-- all covered. Read prac1!
- week 2, Tuesday 12 March
[Unsupervised Classification]
[Snob]
[finite-state]
[2-state]
-- up to
method 1 = method 2 = method 3 - delta
(Later:
[Fisher-info']
[Multi-state]) ...
- week 3, Tuesday 19 March
... use 0- and 1st-order Markov models
[here],
model complexity v. fit to data.
Normal distribution
[N(m,s)]
[Mixtures] --
N(m,s), probability density, -loge prob' density,
data measurement accuracy, mixture of 2 distributions,
separation, uni- or multi-modal,
mixture modelling, total assignment, fractional assignment, search.
Read prac 2!
- week 4, Tuesday 26 March
Discussion re prac 1, due soon.
[Supervised Classification]
[decision trees 1]
firstly binary attributes, topology, split attributes, leaf distributions,
data|tree.
[prac 1] due Thursday 28 March
- week 5, Tuesday 9 April
[decision trees 2]
[an application]
[codes for integers]
- week 6, Tuesday 16 April
Fisher information, message length claculations
[Fisher-info']
[Multi-state],
refer also
[N(mu,s) Fisher].
[Integers]
esp' log*.
[Coding]
inc' sketch of arithmetic coding.
- week 7, Tuesday 23 April
revision
[
paper] (after the event!)
- Examination:
- Date: Monday, 29 April 2002
- Time: 3 - 5pm
- Venue: E2, Building 32
- week 8, see
[CSE455]
Learning and Prediction II
© L. Allison,
School of Computer Science and Software Engineering,
Monash University, Australia 3800.
Created with "vi (Linux & IRIX)", charset=iso-8859-1