Operationalizing Machine Learning

Home

Discover

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and Machine Learning and more.Subscribe

Added to collectionMaybe Cool

CC BY

Source arXiv

Machine Learning

Human-Computer Interaction

Software Engineering

Operationalizing Machine Learning

Operationalizing Machine Learning: An Interview Study

Operationalizing machine learning (mle) pipelines in production requires a continual loop of (i) data collection and labeling, (ii) experimentation to improve ml performance, (iii)evaluation throughout a multi-staged deployment process, and (iv) monitoring of performance drops in production.

We conducted semi-structured ethnographic interviews with 18 machine learning engineers workingacross many applications, including chatbots, autonomous vehicles, and finance.

Our interviews expose three variables that govern success for a production mledeployment : velocity, validation, and versioning.

We summarize common practices for successful mle experimentation, deployment, and sustaining production performance.

We discuss interviewees'pain points and anti-patterns, with implications for tool design.

Authors

Shreya Shankar, Rolando Garcia, Joseph M. Hellerstein, Aditya G. Parameswaran

◐ Recommended Follows

◐ Latest Activity