Biases in Language Models
Diversity & Inclusion, Ethics (Privacy, Fairness,… ), Natural Language Processing

Study of gender biases in popular language models and debiasing model techniques

Do I need to be Dr. Frankenstein to create real-ish synthetic data?
Data Engineering, Ethics (Privacy, Fairness,… ), Governance

Synthetic data not only address the privacy needs but also offer workaround for unprecedented situations. This talk introduces their different types, the options for their generation, and how you don't need to be a mad scientist to make realistic synthetic data

The Myth of Neutrality: How AI is widening social divides
Stefanie Stoppel
Ethics (Privacy, Fairness,… ), Neural Networks / Deep Learning

AI is not neutral and its creation often perpetuates harmful biases. My talk highlights how difficult it is to build "fair and responsible" AI, but also why it's worth to try & prevent these algorithms from cementing existing injustices.