Weakly Supervised Learning via Snorkel - Xinzi Wu

Deep learning is known to be hungry for training data. But high-quality labeled training data are often difficult and expensive to acquire. This talk will give a brief introduction to weakly supervised learning, and demonstrate how to use Snorkel to quickly generate a large amount of probabilistic training labels, which can then be used to train down-stream ML models, including deep learning models.

Xinzi is a principal data scientist with Proofpoint, a cybersecurity company with the mission to protect people, data, and brands. Prior to joining Proofpoint, Xinzi was a senior data scientist at 3M, where she applied both classical machine learning and deep learning to predictive analytics and NLP. Xinzi received a B.S. in computer science from the University of Utah and a Ph.D. in psychology from the University of Virginia.

Location is provided by HireVue
Food will be provided by Google Cloud Platform