Algorithmic bias - Revision history

Lê Nguyên Hoang at 15:50, 23 January 2020

2020-01-23T15:50:06Z

Lê Nguyên Hoang at 12:38, 21 January 2020

2020-01-21T12:38:52Z

Lê Nguyên Hoang at 12:37, 21 January 2020

2020-01-21T12:37:38Z

Lê Nguyên Hoang: Created page with "[https://arxiv.org/pdf/1905.09866.pdf NNG][https://dblp.uni-trier.de/rec/bibtex/journals/corr/abs-1905-09866 19] show that the highly publicized "doctor-man+woman=nurse" is ac..."

2020-01-21T10:39:07Z

Created page with "[https://arxiv.org/pdf/1905.09866.pdf NNG][https://dblp.uni-trier.de/rec/bibtex/journals/corr/abs-1905-09866 19] show that the highly publicized "doctor-man+woman=nurse" is ac..."

New page

[https://arxiv.org/pdf/1905.09866.pdf NNG][https://dblp.uni-trier.de/rec/bibtex/journals/corr/abs-1905-09866 19] show that the highly publicized "doctor-man+woman=nurse" is actually an artefact due to forbidding the use of "doctor" as a reply.

← Older revision		Revision as of 15:50, 23 January 2020
Line 1:		Line 1:
	An algorithmic bias is an (undesirable) bias of an algorithm. In machine learning, this can typically occur if the training dataset contains biased data, e.g. data with historical gender or racial biaises.		An algorithmic bias is an (undesirable) bias of an algorithm. In machine learning, this can typically occur if the training dataset contains biased data, e.g. data with historical gender or racial biaises.
		+
		+	== Impossibility theorem ==
		+
		+	Group and individual fairness are incompatible [https://www.chrisstucchio.com/pubs/slides/crunchconf_2018/slides.pdf Stucchio18].

	== Word embedding ==		== Word embedding ==

@@ Line 5: / Line 5: @@
 The case of word embedding is particularly important, as algorithms rely more and more on natural language processing trained with historical texts. Such texts usually contain a lot of implicit biases which are essentially impossible to clean.
-[https://arxiv.org/pdf/1607.06520.pdf BCZSK][https://dblp.org/rec/bibtex/conf/nips/BolukbasiCZSK16 16] showed that the word embedding of occupations correlated with gender. They found out that "computer programmer - man + woman ≈ homemaker", among other disturbing results.
+[https://arxiv.org/pdf/1607.06520.pdf BCZSK][https://dblp.org/rec/bibtex/conf/nips/BolukbasiCZSK16 16] [https://www.pnas.org/content/pnas/115/16/E3635.full.pdf GSJZ][https://scholar.google.ch/scholar?hl=en&as_sdt=0%2C5&q=Word+embeddings+quantify+100+years+of+gender+and+ethnic+stereotypes&btnG= 18] showed that the word embedding of occupations correlated with gender. They found out that "computer programmer - man + woman ≈ homemaker", among other disturbing results.
 Note however that [https://arxiv.org/pdf/1905.09866.pdf NNG][https://dblp.uni-trier.de/rec/bibtex/journals/corr/abs-1905-09866 19] show that the highly publicized "doctor-man+woman=nurse" is actually an artefact due to forbidding the use of "doctor" as a reply.

← Older revision		Revision as of 12:37, 21 January 2020
Line 1:		Line 1:
−	[https://arxiv.org/pdf/1905.09866.pdf NNG][https://dblp.uni-trier.de/rec/bibtex/journals/corr/abs-1905-09866 19] show that the highly publicized "doctor-man+woman=nurse" is actually an artefact due to forbidding the use of "doctor" as a reply.	+	An algorithmic bias is an (undesirable) bias of an algorithm. In machine learning, this can typically occur if the training dataset contains biased data, e.g. data with historical gender or racial biaises.
		+
		+	== Word embedding ==
		+
		+	The case of word embedding is particularly important, as algorithms rely more and more on natural language processing trained with historical texts. Such texts usually contain a lot of implicit biases which are essentially impossible to clean.
		+
		+	[https://arxiv.org/pdf/1607.06520.pdf BCZSK][https://dblp.org/rec/bibtex/conf/nips/BolukbasiCZSK16 16] showed that the word embedding of occupations correlated with gender. They found out that "computer programmer - man + woman ≈ homemaker", among other disturbing results.
		+
		+	Note however that [https://arxiv.org/pdf/1905.09866.pdf NNG][https://dblp.uni-trier.de/rec/bibtex/journals/corr/abs-1905-09866 19] show that the highly publicized "doctor-man+woman=nurse" is actually an artefact due to forbidding the use of "doctor" as a reply.