Metaheuristics
Metaheuristics are algorithms used in optimization problems. As an
input they usually take reasearch time limit and an initial
solution. With various operations like mutations, choosing an
neighborhood, crossovers etc., changing initial solution more or
less randomly. During research they estimate quality of
encountered candidates. The most popular calssifaction is to
divide them into single-solution and population-based. In
optimization problems we encounter problem as local optimum. There
are some tricks used in Metaheuristics that allow us to escape
from this places. In matehmatical optimization and computer
science they may provide a sufficiently good solution to an
optimization problem. The problem with them is that they do not
guarantee globally optimal solution for some classes of problems.
Some of my implementations can be found
here.
Machine Learning
If machine learning is an aspect of artificial intelligence, then
deep learning
is an aspect of machine learning — furthermore, it is a form of
machine learning that applies
neural networks. I have worked few times with neural networks and linear
regression. Neural networks can be used to solve a lot of
problems. For example we can solve a lot of exploration problems
by learning neural network to predict future data basing on the
data you have. Neural networks can also solve a lot of recognition
problems like object detection on image, or camera. Can be used
for clustering problem in directed and undirected graphs. We could
say the possibilities are endless. Some interesting uses can be
found on my
github.
Other
As long as I devote myself to work and study, I do not really have
a lot of free time. One of my dreams is to conquer one of the
world's greatest peaks, I also like to play computer games :)!
