Disembodied Machine Learning: On the Illusion of Objectivity in NLP
Research output: Contribution to journal › Journal article › Research
Standard
Disembodied Machine Learning: On the Illusion of Objectivity in NLP. / Waseem, Zeerak; Lulz, Smarika ; Bingel, Joachim; Augenstein, Isabelle.
In: OpenReview.net, 2020.Research output: Contribution to journal › Journal article › Research
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - JOUR
T1 - Disembodied Machine Learning: On the Illusion of Objectivity in NLP
AU - Waseem, Zeerak
AU - Lulz, Smarika
AU - Bingel, Joachim
AU - Augenstein, Isabelle
PY - 2020
Y1 - 2020
N2 - Machine Learning (ML) seeks to identify and encode bodies of knowledge within provided datasets. However, data encodes subjective content, which determines the possible outcomes of the models trained on it. Because such subjectivity potentially enables marginalisation of parts of society, it is termed (social) `bias' and sought to be removed. In this opinion paper, we contextualise this discourse of bias in the ML community against the subjective choices in the development process. Through a consideration of how choices in data and model development construct subjectivity, or biases that are represented in a model, we argue that addressing and mitigating biases is near-impossible. This is because both data and ML models are objects for which meaning is made in each step of the development pipeline, from data selection over annotation to model training and analysis. Accordingly, we find the prevalent discourse of bias limiting in its ability to address social marginalisation. We recommend to be conscientious of this, and to accept that de-biasing methods only correct for a fraction of biases.
AB - Machine Learning (ML) seeks to identify and encode bodies of knowledge within provided datasets. However, data encodes subjective content, which determines the possible outcomes of the models trained on it. Because such subjectivity potentially enables marginalisation of parts of society, it is termed (social) `bias' and sought to be removed. In this opinion paper, we contextualise this discourse of bias in the ML community against the subjective choices in the development process. Through a consideration of how choices in data and model development construct subjectivity, or biases that are represented in a model, we argue that addressing and mitigating biases is near-impossible. This is because both data and ML models are objects for which meaning is made in each step of the development pipeline, from data selection over annotation to model training and analysis. Accordingly, we find the prevalent discourse of bias limiting in its ability to address social marginalisation. We recommend to be conscientious of this, and to accept that de-biasing methods only correct for a fraction of biases.
M3 - Journal article
JO - OpenReview.net
JF - OpenReview.net
ER -
ID: 255049388