The starting points for my research have been the Self-Organising Map
(SOM), which has been developed by Professor Teuvo Kohonen, and
principal component analysis (PCA) methods, which has been studied by
professor Erkki Oja.
(If you want to know more about SOM or PCA, see the master's thesis of Jaakko Hollmén for an introduction.)
From June 94 till May 95 I was doing simulations for Kohonen, who was studying ASSOM feature extracting algorithm, which is designed to find invariant features. It was first published in T. Kohonen, Self-Organising Maps, 1995. Translation, rotation and zoom invariant features where studied in article (202 kb).
In May 96 I finished my master's thesis about sparse coding. It is a scheme for representing information, where a small number of units out of a large pool is used to describe each sample. It combines many good sides of vector quantisation and PCA while avoiding most of their drawbacks. Many people have studied the subject, but the best algorithms have been computationally expensive. The complexity of my algorithm is linearly proportional to the input dimension and to the number of neurons. I also had a paper (196 kb) about sparse coding in ICNN'96.