In many industrial ML applications, feature engineering consumes the lion’s share of time, energy and resources. Deep learning promises to replace feature engineering with models that learn end-to-end from “close-to-reality” data, and has convincingly realized this promise for computer vision and natural language processing. Does this promise apply outside of these domains?
In this talk, we will explore how Stripe has applied deep learning techniques to predict fraud from raw behavioral data. Since fraud detection is a critical business problem for Stripe, a well-tuned feature-engineered model is available for comparison. We find that the deep learning model outperforms the feature-engineered model both on predictive performance, and in the effort spent on data engineering, model construction, tuning, and maintenance. Hopefully, this talk will continue to excite a shift in industry practice, towards deeper models trained end-to-end and away from labor-intensive feature engineering.
Pamela Vagata is the AI tech-lead at Stripe, focusing on building deep-learning models. Before joining Stripe, she was a member of Facebook AI Research, developed FBLearner Flow, Facebook’s production ML infrastructure and spent time building data infrastructure.
Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?
Join the conversation here (requires login)
©2019, O'Reilly Media, Inc. • (800) 889-8969 or (707) 827-7019 • Monday-Friday 7:30am-5pm PT • All trademarks and registered trademarks appearing on oreilly.com are the property of their respective owners. • firstname.lastname@example.org