Bias in machine learning - what is it good for?

In public media as well as in scientific publications, the term bias is used in conjunction with machine learning in many different contexts, and with many different meanings. This paper proposes a taxonomy of these different meanings, terminology, and definitions by surveying the, primarily scientific, literature on machine learning. In some cases, we suggest extensions and modifications to promote a clear terminology and completeness. The survey is followed by an analysis and discussion on how different types of biases are connected and depend on each other. We conclude that there is a complex relation between bias occurring in the machine learning pipeline that leads to a model, and the eventual bias of the model (which is typically related to social discrimination). The former bias may or may not influence the latter, in a sometimes bad, and sometime good way.

Dignum, Virginia,ProfessorUmeå universitet,Institutionen för datavetenskap(Swepub:umu)vidi0004 (författare)
Bensch, SunaUmeå universitet,Institutionen för datavetenskap(Swepub:umu)sube0016 (författare)
Umeå universitetInstitutionen för datavetenskap (creator_code:org_t)

Ingår i:NeHuAI 2020 : First International Workshop on New Foundations for Human-Centered AI: RWTH Aachen University, s. 3-10

Av författaren/redakt...: Hellström, Thoma ...; Dignum, Virginia ...; Bensch, Suna

Om ämnet

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.