Title: Introduction - Neural Computation and Self Organization Author: Thomas Breuel Institution: UniKL
from IPython.core.display import Image
def fig(x): return Image(filename="Figures/"+x+".png")
from pylab import *
def figs(*args):
for i,f in enumerate(args):
subplot(1,len(args),i+1)
axis('off')
imshow(imread("Figures/"+f+".png"),cmap=cm.gray)
(Home Page)
(Exercise Sessions)
TBA
Handing in Exercises is mandatory.
Bring your completed exercises to your oral exam.
(Python)
All practical exercises are in Python using iPython Notebooks.
(About Python)
Dynamic, object-oriented programming language.
Python is widely used for scripting, desktop apps, web apps, scientific and numerics, graphics, and games.
Companies making extensive use of Python: Google, Microsoft Research, Yahoo Maps/Groups, Blender, Civ 4, Industrial Light and Magic, Walt Disney Animation, NASA...
Multiple implementations: Python, IronPython (Python-CLR, CE), Cython, PyPy (JIT), jython (Python-JVM)
# Python example: quicksort
from random import choice,sample
def quicksort(l):
if len(l)<2: return l
pivot = choice(l)
lo = [x for x in l if x<pivot]
eq = [x for x in l if x==pivot]
hi = [x for x in l if x>pivot]
return quicksort(lo)+eq+quicksort(hi)
l = sample(range(1000),10)
print l
print quicksort(l)
[60, 240, 70, 754, 27, 67, 786, 57, 100, 965] [27, 57, 60, 67, 70, 100, 240, 754, 786, 965]
(Python properties)
## iPython Notebooks
fig("ipython-notebook")
(Learning Python)
(Important Software)
(course contents)
Introductions to:
(big computer science questions)
# Human Brain Project
figs("hbp")
Brain simulation: http://www.humanbrainproject.eu/index.html
EU 100 million per year
(Human Brain Project)
Life sciences:
Information Technology:
# Obama's BRAIN Initiative
figs("obama-brain")
BRAIN Initiative: http://www.whitehouse.gov/blog/2013/04/02/brain-initiative-challenges-researchers-unlock-mysteries-human-mind
It's a graduate course. It's intended to get you thinking about research and "big picture" questions.
We touch on many topics, but you need to deepen your understanding through reading.
Neural Computation
Supervised Learning
Unsupervised Learning
Also:
Goals:
NB: ML with neural networks is very similar to ML with support vector machines and other approaches