"Personal Research Funding" project PUT1476
PUT1476 "Bridging biological and artificial models of vision (1.01.2017−31.12.2019)", Raul Vicente Zafra, University of Tartu, Faculty of Science and Technology, Institute of Computer Science.
PUT1476
Bioloogiliste ja tehislike nägemistaju mudelite ühendamine
Bridging biological and artificial models of vision
1.01.2017
31.12.2019
R&D project
Personal Research Funding
Exploratory project
Field of researchSubfieldCERCS specialityFrascati Manual specialityPercent
4. Natural Sciences and Engineering4.6. Computer SciencesP176 Artificial intelligence 1.1. Mathematics and computer sciences [mathematics and other allied fields: computer sciences and other allied subjects (software development only hardware development should be classified in the engineering fields)]50,0
4. Natural Sciences and Engineering4.6. Computer SciencesP170 Computer science, numerical analysis, systems, control 1.1. Mathematics and computer sciences [mathematics and other allied fields: computer sciences and other allied subjects (software development only hardware development should be classified in the engineering fields)]50,0
PeriodSum
01.01.2017−31.12.201763 600,00 EUR
01.01.2018−31.12.201863 600,00 EUR
01.01.2019−31.12.201963 600,00 EUR
190 800,00 EUR

Sügavõppe laialdane kasutuselevõtt on viinud oluliste edusammudeni masinnägemises. Samas on inimese nägemisvõime jätkuvalt tehislikest algoritmidest märgatavalt paindlikum. Samuti on siiani ebaselge, kuidas inimnägemine ja sügavad tehisnärvivõrgud üksteisega suhestuvad. Käesoleva projekti eesmärgiks on vähendada masinnägemise ja inimnägemise erinevusi, kandes nende valdkondade vahel üle teadmisi ja arvutuslikke lähenemisi. Selleks kasutame unikaalseid intrakraniaalseid andmeid, võrdlemaks bioloogilist visuaalset töötlust arvutuslike nägemissüsteemidega. Projekti eesmärkideks on tuvastada, kus ja kuidas toimuvad ajus visuaalse signaalitöötluse erinevad etapid, võrrelda masinnägemise õpialgoritme aju õppimisreeglitega, ning uurida, kuidas varasemad kogemused parendavad nägemistaju inimajus ja tehisnärvivõrkudes. Nende küsimuste lahendamine on oluline mõistmaks nägemistaju tööprintsiipe ajus ning võimaldab arendada inimesele sarnaseid nägemisalgoritme.
Fueled by developments in deep learning computer vision has recently achieved spectacular improvements. At the same time human vision still holds important lessons for computer systems to achieve generalization and robustness. Despite their successes it is unknown how human and artificial neural networks for vision relate to each other. The target of this project is to narrow this gap by transferring knowledge and computational strategies between biological and artificial systems of vision. To that end, we will capitalize on unique properties of intracranial brain recordings to compare biological and deep learning systems for vision. In particular, we will determine the neural correlates of hierarchical visual processing, compare neural plasticity rules with optimization algorithms, and explore how prior knowledge can enhance natural and artificial vision. Resolving these issues is important to understand how vision occurs in the brain and to produce human-like vision algorithms.