This year I received an award given out to the students with the highest grade average of the past year. I was honored to receive the award, of course, but I was also delighted about the book I got (where they put the certificate as the first page). In an incredible coincidence the book awarded — Richard Dawkin’s “The Selfish Gene” — was exactly the book I was currently reading on my e-book reader. Although I was halfway through the book at the time, it was a joy to have the rare experience of reading a book in paper format. A full review of the book will be coming soon.
Presented a short paper on machine learning algorithms at this year’s Information Society multiconference. It was a continuation of a project for my Machine Learning course. Prof. Bosnić and I looked at which feature selection techniques and which machine learning algorithms work best for gene microarray data, which has very few observations and many features (genes). The most interesting finding was that genes that were predictive of one cancer were also predictive on other data sets with different types of cancer. Our paper can be found in the proceedings under the Intelligent Systems section (Volume A, pp. 17). More info at my research page.
I just finished reading Steven Pinker’s The Better Angels of Our Nature: Why Violence Has Declined. He presents his viewpoint that the rate of violence has been steadily declining for as long as we have records of it. Being a well-respected professor and researcher, he references a plethora of sources, of almost unimaginable extensiveness, to back up his every claim. From reduction of wars between nations, to civil wars, to homicides, and even to child corporal punishment, he shows that our world has indeed become more peaceful through the ages. Moreover, he brings mathematical rigor to the table and quantifies the probability of starting a war, conflict escalation, as well as duration, among other violent incidents, such as terrorist attacks. All this strengthens his claim and leaves little room for pessimists and their gloomy outlook of the future. I recommend it to anyone who thinks otherwise, and if you need more convincing, you can also check out Bill Gates’s excellent review.
A bit overdue, but (the bulk of) the work I did for my thesis was finally presented at this year’s IEEE ERK (Electrotechnical and Computer Science) conference at Portorož, Slovenia. I did my bachelor’s thesis on unsupervised image segmentation under prof. Matej Kristan of the Visual Cognitive Systems Laboratory at FRI. After improving our approach, we decided to present it at ERK. I was a bit nervous since this was my first oral presentation at a conference, but it went great and the discussion was helpful and interesting. You may find our paper: A regularization-based approach for unsupervised image segmentation Dimitriev Aleksandar, FRI, Uni Lj Kristan Matej, FRI, Uni Lj under the Pattern Recognition section. More details, as well as my other research, can be found at my research page.
Welcome to my site! My name is Aleksandar Dimitriev and I am a computer science student with a penchant for machine learning. This is both my personal and professional website, so you may find not only posts related to ML, but also to science or any topic in general. Feel free to leave any feedback.