Temporal slowness as an unsupervised learning principle [Elektronische Ressource] : self-organization of complex cell receptive fields and application to pattern recognition / von Pietro Berkes

Temporalslownessasanunsupervisedlearningprinciple:self organizationofcomplex cellreceptivefieldsandapplicationtopatternrecognitionDISSERTATIONzurErlangungdesakademischenGradesdoctorrerumnaturalium(Dr.rer.nat.)imFachBiophysikeingereichtanderMathematisch NaturwissenschaftlichenFakultätIderHumboldt UniversitätzuBerlinvonHerrnDipl. Math.PietroBerkesgeborenam4.5.1977inMendrisio(CH)PräsidentderHumboldt UniversitätzuBerlin:Prof.Dr.JürgenMlynekDekanderMathematisch NaturwissenschaftlichenFakultätI:Prof.ThomasBuckhout,PhDGutachter:1. LaurenzWiskott2. PeterKönig3. AndreasHerzeingereichtam: 30.3.2005TagdermündlichenPrüfung: 23.6.2005IIAcknowledgmentsI am aware that this work would not have been possible without the help of many people, to which I amgreatlyindebted. Myfirstthoughtgoestomyadvisor,LaurenzWiskott,whosemathematicalintuitionneverceasedtoamazeme. Intheseyearshetaughtmetoomanythingstomention,mostlybyhisexample,andIamextremelygratefulforthis. IhopeIwillbeabletoworkwithyouagain,sometimeinthefuture.TobiasBlaschkewasalreadyworkinginLaurenz’groupwhenIfirstarrived. Heintroducedmebothtothe secrets of the ITB and to Berlin. Thank you for all the technical discussions we had and even more foralotofinsidertipsaboutBerlinnightlife. SpecialthankstoTizianoZito,colleagueandfriend,withwhomI wrote a nice piece of software (MDP) and heard the guitar of Max Cavalera, live at the Columbiahalle.