Explore the intricacies of designing machine learning products and systems in this insightful webinar featuring Grace Huang, Data Science Manager at Pinterest. Delve into crucial considerations beyond algorithm implementation, including handling shifting data landscapes, maintaining a healthy data ecosystem, and addressing bias. Learn valuable lessons on creating ML-minded products and product-minded ML systems, covering topics such as data considerations, evaluation metrics, product monitoring, automation, and bias detection. Gain practical insights on testing, monitoring, infrastructure, measurement, and troubleshooting ML systems. Discover when ML is truly necessary and how to effectively integrate it into real-world applications, drawing from Huang's extensive experience in recommendation systems, search relevance, and algorithm development across various fields.
Designing an ML Minded Product and a Product Minded ML System
Association for Computing Machinery (ACM) via YouTube
Overview
Syllabus
Introduction
Grace Huang Introduction
Before You Start
Agenda
Data Considerations
User Data
Logging
Poll
Training Data
Evaluation
Metrics
Product Monitoring
Automation
Data System Bias
Bias in ML Models
Testing and Monitoring
Infrastructure
Measurement and understanding
Troubleshooting
Do you really need ML
Taught by
Association for Computing Machinery (ACM)
