Naslov
Computational Complexity, Machine Learning, and Philosophy of Mind: Preliminary Reports

Autori
Šekrst, Kristina

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Skup
Formal Methods and Science in Philosophy

Mjesto i datum
Dubrovnik, Hrvatska, 26.03.2015. - 28.03.2015

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
computational complexity ; machine learning ; neural networks ; P vs. NP ; artificial intelligence

Sažetak
Computational complexity theory is a branch of computer science and mathematics, which deals with classification and analysis of various computational problems regarding the amount of resources one must use to solve a problem or to verify it. The P versus NP problem is a major unsolved problem in modern mathematics and computer science, which deals with the question whether a problem can be solved quickly using a computer if one can verify it quickly. P problems are ones that run in polynomial time and a computer can can rapidly provide us with an answer, while NP ones are characterized by the ability to verify the answer quickly, but not providing us with one in the same amount of time (which often rises exponentially or worse). If one can state that NP problems can be reduced to P problems, we would be able to solve these hard problems quickly, which would mean a great advance in philosophy of mind and artificial intelligence. In this lecture, we will observe some major paradigms in machine learning, such as clustering and neural networks, especially regarding unsupervised learning and fast algorithms, which try to simulate polynomial time solving, and that will lead us to possible explanations of certain mental phenomena and a possible way of a path for resolving burning issues in philosophy of mind.

Izvorni jezik
Engleski

Znanstvena područja
Matematika, Računarstvo, Filozofija

POVEZANOST RADA