prikaz prve stranice dokumenta Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja
Access restricted to authenticated users
master's thesis
Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja
Split: University of Split, University of Split, Faculty of science, 2020. urn:nbn:hr:166:335297
Vrdoljak, Ema
University of Split
Faculty of Science
Department of Informatics

Institutional repository: Repository of Faculty of Science

Cite this document

Vrdoljak, E. (2020). Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja (Master's thesis). Split: University of Split, Faculty of Science. Retrieved from https://urn.nsk.hr/urn:nbn:hr:166:335297

Vrdoljak, Ema. "Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja." Master's thesis, University of Split, Faculty of Science, 2020. https://urn.nsk.hr/urn:nbn:hr:166:335297

Vrdoljak, Ema. "Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja." Master's thesis, University of Split, Faculty of Science, 2020. https://urn.nsk.hr/urn:nbn:hr:166:335297

Vrdoljak, E. (2020). 'Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja', Master's thesis, University of Split, Faculty of Science, accessed 30 July 2024, https://urn.nsk.hr/urn:nbn:hr:166:335297

Vrdoljak E. Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja [Master's thesis]. Split: University of Split, Faculty of Science; 2020 [cited 2024 July 30] Available at: https://urn.nsk.hr/urn:nbn:hr:166:335297

E. Vrdoljak, "Kvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja", Master's thesis, University of Split, Faculty of Science, Split, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:166:335297

Metadata
TitleKvantitativno vrednovanje raspodijeljenog inkrementalnog strojnog učenja
Title (english)Quantitative evaluation of distributed online machine learning
AuthorEma Vrdoljak
MentorMarko Rosić (mentor)
Committee memberMarko Rosić (predsjednik povjerenstva)
Committee memberMarin Aglić Čuvić (član povjerenstva)
Committee memberGoran Zaharija (član povjerenstva)
GranterUniversity of Split
Faculty of Science
(Department of Informatics)
Split
Defense date and country2020-09-23, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U inkrementalnom raspodijeljenom strojnom učenju treniranje modela se vrši nad tokovima podataka čiji se izvori nalaze na fizički udaljenim lokacijama. Kod raspodijeljenih sustava svaki čvor ima vlastiti model koji je potrebno tijekom vremena unaprijediti. Kontinuirani tok resursa, fizički udaljeni čvorovi i ograničeni protok mreže dovode do problema odabira optimalnog vremena ažuriranja modela. U ovom radu opisane su funkcije koje omogućavaju predviđanje ponašanja modela čvorova i njihovih točnosti. Rezultati pokazuju da je prikladno vrijeme ažuriranja svih modela nakon primitka odgovora od drugog čvora. Treba uzeti u obzir da se optimalno vrijeme razlikuje ovisno o situaciji i da rezultati dobiveni u ovom radu nisu primjenjivi na ostale probleme.
Abstract (english)
In distributed online machine learning, data streams are use to perform training model. Data sources are located at physically remote locations. In distributed systems, each node has its own model that needs to be improved over time. Continuously resource flow, physically remote nodes and limited network flow lead to the problem of selecting the optimal update time of the model. This paper describes the functions that predicting the behaviour of nodes models and their accuracy. The results show that appropriate update time of the all models is after receiving a response from the second node. It should be noted that the optimal time varies regardless of the situation and that the result obtained in this paper are not applicable to other problems.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:166:335297
Study programmeTitle: Computer Science; specializations in: major in teaching
Course: major in teaching
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra edukacije informatike (magistar/magistra edukacije informatike)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2021-01-13 17:39:25