Palestra: A look at the methods to detect and try to remove bias in Machine Learning models

Track: Machine Learning e Inteligência Artificial

Sala: Sala 02

Horário: 11:50am - 12:35pm

Dia da semana: Terça-feira

Nível: Intermediário - Avançado

Persona: Cientista de Dados

Apresentação em Inglês

Compartilhe essa página no:

Pontos Principais

  • How to be fair when using machine learning;
  • Example of common pitfalls;
  • Pros and Cons of different methods

Resumo

Machine Learning is impacting a lot of aspect of society and people often forget how their choices are impacting the lives of people on the other end of the prediction. Today, more and more tools are available to help data scientists do their jobs.

In this talk, we will explore some examples where machine learning fails and/or is making a negative impact. We will look at some of the tools available today to try checking if the model we create are discriminatory. Being fair shouldn’t mean losing value, on the contrary.

Palestrante: Thierry Silbermann

Senior Data Scientist at Nubank

Thierry did a double Master degree in Computer Science from ESIEA (French Engineer School) in France and the Illinois Institute of Technology in Chicago, USA. After that he worked for a startup in Paris for 6 months before starting a PhD at the University of Konstanz in Germany on Recommendation System working on libFM. Today he works as a Tech Lead Data Scientist on different kinds of project ranging from improving customer service, understanding growth, improving processes and managing other data scientists. He is also the co-organizer of the Machine Learning Meetup in Sao Paulo which is the biggest in Latin America.

Encontre mais informações sobre Thierry Silbermann em

Tracks

Segunda-feira, 6 de maio

Terça-feira, 7 de maio

Quarta-feira, 8 de maio