When design meets machine learning
summary
Design creates interfaces that feed ML actionable signals and offers four practical principles: explicitness frequency specificity consistency. Quora’s Distribution work targets 300+ million users and shows examples: bookmarks and likes are explicit but rare, Tinder forces frequent like/dislike input, TikTok and YouTube use watch time for specificity, a missing Read More capture biases models, and Instagram enforces consistent image sizes so time-on-screen compares fairly. These principles were iterated over months and deployed.
AI-written summary · about
chapters (10)
- 0:00Introduction
- 2:24How do we design a product that learns from its users
- 3:06What is Machine Learning?
- 7:03Design enables ML by creating interfaces that capture the user’s preferences
- 7:33Signals
- 8:38Explicitness
- 12:06Frequency
- 14:19Specificity
- 17:22Consistency
- 20:33Conclusion
related talks
25mDesign for Context with Machine Learning
27mContinous integration of ML products and UX design
12mHow we infuse our biases in the products...
44mHow to Improve UX with Machine Learning: Insights and learnings until now
38mDesigning Machine Learning for Humans
20mMicrointeractions for the System: Design in the Age of Techlash
28mDesigning for Impact: Harnessing Collaboration & Innovation
26mDesigning Anticipated User Experiences