- author = "Rafal Salustowicz",
- title = "Probabilistic Incremental Program Evolution",
- school = "Fakultaet IV - Elektrotechnik und Informatik, Technical University of Berlin",
- year = "2003",
- address = "Germany",
- month = "3 " # apr # " 2003",
- keywords = "genetic algorithms, genetic programming, PIPE, Long Short-Term Memory, LSTM, evolutionary optimization algorithm, Evolutionaerer Optimierungs, Algorithmus, Programm Evolution, Stochastische Programier Sprachen, Stochastische Programm Suche, Evolutionary Optimization Algorithm, Probabilistic Programing Languages, Program Evolution, Stochastic Program Search",
- URN = "urn:nbn:de:kobv:83-opus-6503",
- timestamp = "Sat, 17 Jul 2021 09:03:10 +0200",
-
biburl = "
https://dblp.org/rec/phd/dnb/Salustowicz03.bib",
- bibsource = "dblp computer science bibliography, https://dblp.org",
-
URL = "https://depositonce.tu-berlin.de/handle/11303/1046",
-
URL = "https://api-depositonce.tu-berlin.de/server/api/core/bitstreams/512a880f-2c97-4af1-bd47-a83e72aa4627/content",
-
DOI = "
doi:10.14279/depositonce-749",
- size = "157 pages",
- abstract = "Probabilistic Incremental Program Evolution (PIPE) is a machine learning (ML) technique. Just like other ML techniques such as, e.g., neural networks, reinforcement learning, or evolutionary algorithms, PIPE tries to enable computers to solve problems automatically, i.e. to find solutions by learning from experience (examples), rather than being explicitly programmed to solve a task. PIPE is an evolutionary optimization algorithm, which employs stochastic models to search for computer programs that embody solutions to given problems.",
-
notes = "in English
The work described in this thesis was performed at IDSIA: Istituto Dalle Molle di Studi sull’Intelligenza Artificiale in Lugano, Switzerland. It was made possible due to funding of IDSIA, and was supported in part by the Swiss National Fund (SNF) grant 2100-49’144.96 Long Short-Term Memory.
Supervisor: Juergen Schmidhuber",
