Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.

Neutralising bias on word embeddings

241 vues

Publié le

The spread of Artificial Intelligence, mostly in the form of representational learning, has introduced an issue that, at first glance, seems difficult to get fixed: bias.

Throughout the last years, more and more examples have emerged where people got mistaken by animals, bots became racists and machine learning solutions, chiefly used by recruitment companies and translation services, spot on the news with bias related issues.

In this presentation, we will see the impact bias has on people and how to fix it without having to dive deep into the data and remove it manually.

Publié dans : Sciences
  • Identifiez-vous pour voir les commentaires

Neutralising bias on word embeddings

  2. 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. 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. 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. 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. 6. –Joanna Bryson, University of Bath and Princeton University "AI is just an extension of our existing culture.”
  7. 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. 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. 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. 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. 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. 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. 13. Identifying gender [woman] - [man] = [female]
  14. 14. What about other words?
  15. 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. 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
  17. 17. Equalising gender-specific words Tricky parts!
  18. 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. 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. 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