Explore the critical issue of bias in artificial intelligence through this keynote speech by Margaret Mitchell, PhD, delivered at the ODSC East 2020 Virtual Conference. Delve into the complexities of vision-language AI and grounded language generation, examining how biases in data representation and human cognition can lead to unjust outcomes in AI systems. Learn about the challenges in predictive policing, computer vision, and automated inference, and discover strategies for mitigating bias, including disaggregated evaluation, intersectional analysis, and improved data handling. Gain insights into the importance of privacy, documentation, and inclusive data practices in AI development. Understand the potential of synthetic data and the need for established standards to create more equitable AI systems that evolve towards positive societal goals.
Bias in the Vision and Language of Artificial Intelligence
Overview
Syllabus
Introduction
What do you see
A riddle
Gender norms
Outcomes properties
AI pipeline
Human biases
Biased data representation
Bias network effect
Bias amplifies injustice
Bias in predictive policing
Bias in computer vision
Predicting criminality
Automated inference on criminality
Predicting homosexuality
What now
De disaggregated evaluation
How this works
Intersection
Confusion Matrix
Precision Force
Acceptable tradeoffs
Privacy and images
False negatives
Data constraints
Google Translate
Unjust outcomes
Handle your data
Tools
Documentation
Conclusion
Questions
All data is biased
Inclusive Images
Standards
Synthetic Data
Taught by
Open Data Science
