SlideShare ist ein Scribd-Unternehmen logo
1 von 21
Downloaden Sie, um offline zu lesen
NEUTRALISING BIAS ON WORD
EMBEDDINGS
–Wilder Rodrigues
Wilder Rodrigues
• Machine Learning Engineer at Quby;
• Coursera Mentor;
• City.AI Ambassador;
• School of AI Dean [Utrecht]
• IBM Watson AI XPRIZE contestant;
• Kaggler;
• Public speaker;
• Family man and father of 3.
@wilderrodrigues
https://medium.com/@wilder.rodrigues
How do you see racism?
• Before you proceed, please watch this video: https://www.youtube.com/watch?v=5F_atkP3pqs
• The audio is in Portuguese, but in the next slide you will find translations for what people said in the
interviews.
Source: Canal deTV da FAP (Astrojildo Pereira Foundation)
Translations
• Group 1
• He is late;
• She is a fashion designer;
• Holds an executive position in either the HR
or Finance area;
• Taking care of his garden. Doesn’t look like a
gardener;
• She is cleaning her own house; the countertop;
• Graffiti artist; it’s an art, it’s not vandalism.
• Group II
• Vandalising the wall; she is a spitter;
• She is a housekeeper; cleaning the house;
• He is a gardener;
• He looks like a security guard or a
chauffeur;
• Seamstress; saleswoman;
• He is running away; he is a thief.
Unconscious bias
• Blue is for boys, pink for girls.
• Boys are better at maths and science.
• Tall people make better leaders.
• New mothers are more absent from work
than new fathers.
• People with tattoos are rebellious.
• Younger people are better with technology
than older people.
–Joanna Bryson, University of Bath and Princeton University
"AI is just an extension of our existing culture.”
Racialized code & Unregulated algorithms
Source: https://www.theguardian.com/technology/2017/dec/04/racist-facial-recognition-white-coders-black-people-police
Joy Buolamwini, Code4Rights and MIT Media Lab Researcher.
How white engineers built racist code – and
why it's dangerous for black people
Source: https://www.theguardian.com/technology/2017/dec/04/racist-facial-recognition-white-coders-black-people-police
Implicit AssociationTest
Both black and white Americans, for
example, are faster at associating names
like “Brad” and “Courtney” with words
like “happy” and “sunrise,” and names like
“Leroy” and “Latisha” with words like
“hatred” and “vomit” than vice versa.
Source: http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
W.E.A.T
Names like “Brett” and “Allison” were
more similar to those for positive words
including love and laughter, and those for
names like “Alonzo” and “Shaniqua” were
more similar to negative words like
“cancer” and “failure.” 
Source: http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
W.E.F.A.T
How closely related the embeddings for
words like “hygienist” and “librarian” were
to those of words like “female” and
“woman.” It then compared this
computer-generated gender association
measure to the actual percentage of
women in that occupation.
Source: http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
Word Embeddings
A ⋅ B
∥A∥∥B∥
=
∑
n
i=1
AiBi
∑
n
i=1
A2
i ∑
n
i=1
B2
i
Source: https://medium.com/cityai/deep-learning-for-natural-language-processing-part-i-8369895ffb98
Father (L2 norm): 5.31
Mother (L2 norm): 5.63
d: 26.67
p: 29.89
Similarity: d / p = 0.89
Car (L2 norm): 5.73
Bird (L2 norm): 4.83
d: 5.96
p: 27.67
Similarity: d / p = 0.21
Identifying gender
[woman] - [man] = [female]
What about other words?
Neutralising bias from non-gender specific
words
ebias_comp
=
e ⋅ g
∥g∥2
2
g
edebiased
= e − ebias
Source: Bolukbasi et al., 2016, https://arxiv.org/pdf/1607.06520.pdf
Does it work?
• Cosine similarity between receptionist
and gender, before neutralising:
• 0.3307794175059373
• Cosine similarity between receptionist
and gender, after neutralising:
• 5.2021694209043796e-17
Equalising gender-specific words
Tricky
parts!
Equalising gender-specific words
• Cosine similarity between actor and gender, before
equalising:
• -0.08387555382505694
• Cosine similarity between actress and gender, before
equalising::
• 0.33422494897899785
• Cosine similarity between actor and gender, after
equalising:
• -0.8796563888581831
• Cosine similarity between actress and gender, after
equalising:
• 0.879656388858183
How far is actor from babysitter?
• Cosine similarity between actor and babysitter, before
neutralising:
• 0.2766562472128601
• Cosine similarity between actress and babysitter, before
neutralising::
• 0.3378475317457311
• Cosine similarity between actor and babysitter, after
neutralising:
• 0.1408988327631711
• Cosine similarity between actress and babysitter, after
neutralising:
• 0.14089883276317122
References
• https://www.youtube.com/watch?v=5F_atkP3pqs
• https://www.theguardian.com/technology/2017/dec/04/racist-facial-recognition-white-coders-black-people-police
• http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
• https://medium.com/cityai/deep-learning-for-natural-language-processing-part-i-8369895ffb98
• Bolukbasi et al., 2016, https://arxiv.org/pdf/1607.06520.pdf
• Jeffrey Pennington, Richard Socher, and Christopher D. Manning, https://nlp.stanford.edu/projects/glove/
• https://github.com/ekholabs/DLinK/blob/master/notebooks/nlp/neutralising-equalising-word-embeddings.ipynb
Neutralising bias on word embeddings

Weitere ähnliche Inhalte

Ähnlich wie Neutralising bias on word embeddings

Harvard Essay Examples.pdf
Harvard Essay Examples.pdfHarvard Essay Examples.pdf
Harvard Essay Examples.pdfAlison Parker
 
Plagiarism
PlagiarismPlagiarism
Plagiarismaislater
 
Essay Body Paragraph Generator
Essay Body Paragraph GeneratorEssay Body Paragraph Generator
Essay Body Paragraph GeneratorMichelle Price
 
Printable Elementary Lined Paper - Printable World Ho
Printable Elementary Lined Paper - Printable World HoPrintable Elementary Lined Paper - Printable World Ho
Printable Elementary Lined Paper - Printable World HoHeidi Perry
 
Outline To An Essay.pdf
Outline To An Essay.pdfOutline To An Essay.pdf
Outline To An Essay.pdfWendy Bolden
 
Cause And Effect Paragraph Ppt. How To Write A Caus
Cause And Effect Paragraph Ppt. How To Write A CausCause And Effect Paragraph Ppt. How To Write A Caus
Cause And Effect Paragraph Ppt. How To Write A CausVeronica Rogers
 

Ähnlich wie Neutralising bias on word embeddings (8)

Harvard Essay Examples.pdf
Harvard Essay Examples.pdfHarvard Essay Examples.pdf
Harvard Essay Examples.pdf
 
Rodriguez irizarry
Rodriguez  irizarryRodriguez  irizarry
Rodriguez irizarry
 
Plagiarism
PlagiarismPlagiarism
Plagiarism
 
Essay Body Paragraph Generator
Essay Body Paragraph GeneratorEssay Body Paragraph Generator
Essay Body Paragraph Generator
 
Printable Elementary Lined Paper - Printable World Ho
Printable Elementary Lined Paper - Printable World HoPrintable Elementary Lined Paper - Printable World Ho
Printable Elementary Lined Paper - Printable World Ho
 
Outline To An Essay.pdf
Outline To An Essay.pdfOutline To An Essay.pdf
Outline To An Essay.pdf
 
Essay For Teachers.pdf
Essay For Teachers.pdfEssay For Teachers.pdf
Essay For Teachers.pdf
 
Cause And Effect Paragraph Ppt. How To Write A Caus
Cause And Effect Paragraph Ppt. How To Write A CausCause And Effect Paragraph Ppt. How To Write A Caus
Cause And Effect Paragraph Ppt. How To Write A Caus
 

Mehr von Wilder Rodrigues

Improving Machine Learning
 Workflows: Training, Packaging and Serving.
Improving  Machine Learning
 Workflows: Training, Packaging and Serving.Improving  Machine Learning
 Workflows: Training, Packaging and Serving.
Improving Machine Learning
 Workflows: Training, Packaging and Serving.Wilder Rodrigues
 
Deep Learning for Natural Language Processing
Deep Learning for Natural Language ProcessingDeep Learning for Natural Language Processing
Deep Learning for Natural Language ProcessingWilder Rodrigues
 
Microservices with Spring Cloud
Microservices with Spring CloudMicroservices with Spring Cloud
Microservices with Spring CloudWilder Rodrigues
 
Embracing Reactive Streams with Java 9 and Spring 5
Embracing Reactive Streams with Java 9 and Spring 5Embracing Reactive Streams with Java 9 and Spring 5
Embracing Reactive Streams with Java 9 and Spring 5Wilder Rodrigues
 

Mehr von Wilder Rodrigues (7)

Improving Machine Learning
 Workflows: Training, Packaging and Serving.
Improving  Machine Learning
 Workflows: Training, Packaging and Serving.Improving  Machine Learning
 Workflows: Training, Packaging and Serving.
Improving Machine Learning
 Workflows: Training, Packaging and Serving.
 
Deep Learning for Natural Language Processing
Deep Learning for Natural Language ProcessingDeep Learning for Natural Language Processing
Deep Learning for Natural Language Processing
 
Ai - A Practical Approach
Ai - A Practical ApproachAi - A Practical Approach
Ai - A Practical Approach
 
Java 9: Jigsaw Project
Java 9: Jigsaw ProjectJava 9: Jigsaw Project
Java 9: Jigsaw Project
 
Microservices with Spring Cloud
Microservices with Spring CloudMicroservices with Spring Cloud
Microservices with Spring Cloud
 
Machine intelligence
Machine intelligenceMachine intelligence
Machine intelligence
 
Embracing Reactive Streams with Java 9 and Spring 5
Embracing Reactive Streams with Java 9 and Spring 5Embracing Reactive Streams with Java 9 and Spring 5
Embracing Reactive Streams with Java 9 and Spring 5
 

Kürzlich hochgeladen

Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .Poonam Aher Patil
 
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls AgencyHire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls AgencySheetal Arora
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLkantirani197
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryAlex Henderson
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Joonhun Lee
 
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...ssuser79fe74
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptxAlMamun560346
 
Nanoparticles synthesis and characterization​ ​
Nanoparticles synthesis and characterization​  ​Nanoparticles synthesis and characterization​  ​
Nanoparticles synthesis and characterization​ ​kaibalyasahoo82800
 
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...Lokesh Kothari
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPirithiRaju
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPirithiRaju
 
GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)Areesha Ahmad
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)Areesha Ahmad
 
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPirithiRaju
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)Areesha Ahmad
 
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxCOST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxFarihaAbdulRasheed
 
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICESAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICEayushi9330
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and ClassificationsAreesha Ahmad
 
TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...
TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...
TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...ssifa0344
 

Kürzlich hochgeladen (20)

Factory Acceptance Test( FAT).pptx .
Factory Acceptance Test( FAT).pptx       .Factory Acceptance Test( FAT).pptx       .
Factory Acceptance Test( FAT).pptx .
 
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls AgencyHire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
Hire 💕 9907093804 Hooghly Call Girls Service Call Girls Agency
 
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRLKochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
Kochi ❤CALL GIRL 84099*07087 ❤CALL GIRLS IN Kochi ESCORT SERVICE❤CALL GIRL
 
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and SpectrometryFAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
FAIRSpectra - Enabling the FAIRification of Spectroscopy and Spectrometry
 
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
Feature-aligned N-BEATS with Sinkhorn divergence (ICLR '24)
 
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
Chemical Tests; flame test, positive and negative ions test Edexcel Internati...
 
Seismic Method Estimate velocity from seismic data.pptx
Seismic Method Estimate velocity from seismic  data.pptxSeismic Method Estimate velocity from seismic  data.pptx
Seismic Method Estimate velocity from seismic data.pptx
 
Nanoparticles synthesis and characterization​ ​
Nanoparticles synthesis and characterization​  ​Nanoparticles synthesis and characterization​  ​
Nanoparticles synthesis and characterization​ ​
 
Site Acceptance Test .
Site Acceptance Test                    .Site Acceptance Test                    .
Site Acceptance Test .
 
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
Labelling Requirements and Label Claims for Dietary Supplements and Recommend...
 
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdfPests of cotton_Sucking_Pests_Dr.UPR.pdf
Pests of cotton_Sucking_Pests_Dr.UPR.pdf
 
Pests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdfPests of mustard_Identification_Management_Dr.UPR.pdf
Pests of mustard_Identification_Management_Dr.UPR.pdf
 
GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)GBSN - Microbiology (Unit 2)
GBSN - Microbiology (Unit 2)
 
GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)GBSN - Microbiology (Unit 3)
GBSN - Microbiology (Unit 3)
 
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdfPests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
Pests of cotton_Borer_Pests_Binomics_Dr.UPR.pdf
 
GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)GBSN - Microbiology (Unit 1)
GBSN - Microbiology (Unit 1)
 
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptxCOST ESTIMATION FOR A RESEARCH PROJECT.pptx
COST ESTIMATION FOR A RESEARCH PROJECT.pptx
 
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICESAMASTIPUR CALL GIRL 7857803690  LOW PRICE  ESCORT SERVICE
SAMASTIPUR CALL GIRL 7857803690 LOW PRICE ESCORT SERVICE
 
Bacterial Identification and Classifications
Bacterial Identification and ClassificationsBacterial Identification and Classifications
Bacterial Identification and Classifications
 
TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...
TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...
TEST BANK For Radiologic Science for Technologists, 12th Edition by Stewart C...
 

Neutralising bias on word embeddings

  • 1. NEUTRALISING BIAS ON WORD EMBEDDINGS –Wilder Rodrigues
  • 2. Wilder Rodrigues • Machine Learning Engineer at Quby; • Coursera Mentor; • City.AI Ambassador; • School of AI Dean [Utrecht] • IBM Watson AI XPRIZE contestant; • Kaggler; • Public speaker; • Family man and father of 3. @wilderrodrigues https://medium.com/@wilder.rodrigues
  • 3. How do you see racism? • Before you proceed, please watch this video: https://www.youtube.com/watch?v=5F_atkP3pqs • The audio is in Portuguese, but in the next slide you will find translations for what people said in the interviews. Source: Canal deTV da FAP (Astrojildo Pereira Foundation)
  • 4. Translations • Group 1 • He is late; • She is a fashion designer; • Holds an executive position in either the HR or Finance area; • Taking care of his garden. Doesn’t look like a gardener; • She is cleaning her own house; the countertop; • Graffiti artist; it’s an art, it’s not vandalism. • Group II • Vandalising the wall; she is a spitter; • She is a housekeeper; cleaning the house; • He is a gardener; • He looks like a security guard or a chauffeur; • Seamstress; saleswoman; • He is running away; he is a thief.
  • 5. Unconscious bias • Blue is for boys, pink for girls. • Boys are better at maths and science. • Tall people make better leaders. • New mothers are more absent from work than new fathers. • People with tattoos are rebellious. • Younger people are better with technology than older people.
  • 6. –Joanna Bryson, University of Bath and Princeton University "AI is just an extension of our existing culture.”
  • 7. Racialized code & Unregulated algorithms Source: https://www.theguardian.com/technology/2017/dec/04/racist-facial-recognition-white-coders-black-people-police Joy Buolamwini, Code4Rights and MIT Media Lab Researcher.
  • 8. How white engineers built racist code – and why it's dangerous for black people Source: https://www.theguardian.com/technology/2017/dec/04/racist-facial-recognition-white-coders-black-people-police
  • 9. Implicit AssociationTest Both black and white Americans, for example, are faster at associating names like “Brad” and “Courtney” with words like “happy” and “sunrise,” and names like “Leroy” and “Latisha” with words like “hatred” and “vomit” than vice versa. Source: http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
  • 10. W.E.A.T Names like “Brett” and “Allison” were more similar to those for positive words including love and laughter, and those for names like “Alonzo” and “Shaniqua” were more similar to negative words like “cancer” and “failure.”  Source: http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
  • 11. W.E.F.A.T How closely related the embeddings for words like “hygienist” and “librarian” were to those of words like “female” and “woman.” It then compared this computer-generated gender association measure to the actual percentage of women in that occupation. Source: http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender
  • 12. Word Embeddings A ⋅ B ∥A∥∥B∥ = ∑ n i=1 AiBi ∑ n i=1 A2 i ∑ n i=1 B2 i Source: https://medium.com/cityai/deep-learning-for-natural-language-processing-part-i-8369895ffb98 Father (L2 norm): 5.31 Mother (L2 norm): 5.63 d: 26.67 p: 29.89 Similarity: d / p = 0.89 Car (L2 norm): 5.73 Bird (L2 norm): 4.83 d: 5.96 p: 27.67 Similarity: d / p = 0.21
  • 13. Identifying gender [woman] - [man] = [female]
  • 15. Neutralising bias from non-gender specific words ebias_comp = e ⋅ g ∥g∥2 2 g edebiased = e − ebias Source: Bolukbasi et al., 2016, https://arxiv.org/pdf/1607.06520.pdf
  • 16. Does it work? • Cosine similarity between receptionist and gender, before neutralising: • 0.3307794175059373 • Cosine similarity between receptionist and gender, after neutralising: • 5.2021694209043796e-17
  • 18. Equalising gender-specific words • Cosine similarity between actor and gender, before equalising: • -0.08387555382505694 • Cosine similarity between actress and gender, before equalising:: • 0.33422494897899785 • Cosine similarity between actor and gender, after equalising: • -0.8796563888581831 • Cosine similarity between actress and gender, after equalising: • 0.879656388858183
  • 19. How far is actor from babysitter? • Cosine similarity between actor and babysitter, before neutralising: • 0.2766562472128601 • Cosine similarity between actress and babysitter, before neutralising:: • 0.3378475317457311 • Cosine similarity between actor and babysitter, after neutralising: • 0.1408988327631711 • Cosine similarity between actress and babysitter, after neutralising: • 0.14089883276317122
  • 20. References • https://www.youtube.com/watch?v=5F_atkP3pqs • https://www.theguardian.com/technology/2017/dec/04/racist-facial-recognition-white-coders-black-people-police • http://www.sciencemag.org/news/2017/04/even-artificial-intelligence-can-acquire-biases-against-race-and-gender • https://medium.com/cityai/deep-learning-for-natural-language-processing-part-i-8369895ffb98 • Bolukbasi et al., 2016, https://arxiv.org/pdf/1607.06520.pdf • Jeffrey Pennington, Richard Socher, and Christopher D. Manning, https://nlp.stanford.edu/projects/glove/ • https://github.com/ekholabs/DLinK/blob/master/notebooks/nlp/neutralising-equalising-word-embeddings.ipynb