ScholarGate
Assistent
Machine learningMachine learning

Bayesiansk aktiv læring

Bayesiansk aktiv læring (BAL) kombinerer en probabilistisk model med en aktiv forespørgselsstrategi for at identificere de umærkede eksempler, som, når de er mærket, mest vil reducere modelusikkerhed. I stedet for at mærke data tilfældigt, styrer BAL en orakel – typisk en menneskelig annotator – mod de punkter, hvor mærkning vil give den største informationsgevinst, hvilket gør den yderst label-effektiv.

Åbn i MethodMindSnartVideoSnartDownload slides

Læs hele metoden

Kun for medlemmer

Log ind med en gratis konto for at læse dette afsnit.

Log ind

Method map

The neighbourhood of related methods — select a node to explore.

Kilder

  1. Houlsby, N., Huszár, F., Ghahramani, Z., & Lengyel, M. (2011). Bayesian Active Learning for Classification and Preference Learning. arXiv preprint arXiv:1112.5745. link
  2. Settles, B. (2012). Active Learning. Synthesis Lectures on Artificial Intelligence and Machine Learning, 6(1), 1–114. Morgan & Claypool. DOI: 10.2200/S00429ED1V01Y201207AIM018

Sådan citerer du denne side

ScholarGate. (2026, June 3). Bayesian Active Learning (Query-by-Committee and BALD). ScholarGate. https://scholargate.app/da/machine-learning/bayesian-active-learning

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

Compare side by side

Refereret af

ScholarGateBayesian Active Learning (Bayesian Active Learning (Query-by-Committee and BALD)). Hentet 2026-06-15 fra https://scholargate.app/da/machine-learning/bayesian-active-learning · Datasæt: https://doi.org/10.5281/zenodo.20539026