Operational Challenges in Machine Learning Model Life Cycle
summary
Walmart Labs builds an Analyzer that instruments user events, stores them in an event repository, tags signals with a signal generator, and models behavior using k-means clustering. The system clusters users into 85 experts, 100 intermediates and 190 beginners, surfaces repeated errors (resource-creation conflicts hit 50% of experts, 65% of intermediates, 80% of beginners) and classifies 70 education/training issues and 6 platform-simplicity issues, which leads to new documentation and targeted UI fixes now underway.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 0:50Agenda
- 1:32Background
- 4:05Problem Statement
- 8:12Architecture
- 10:11Platform
- 10:51Model
- 17:40Results
related talks
23mMachine Learning with a Deep Dive into Automated Forecasting System
30mOperationalizing Machine Learning
42mTesting Machine Learning Modules in Production
30mChallenges of machine learning: The new ML lifecycle
32mReliable Machine Learning
13mOperationalizing machine learning for the enterprise
48mPrinciples and Practices of Building Machine Learning Systems
23mHow Can We Trust Machine Learning? Exploration, Evaluation...