The age of machine learning
summary
Argues there are three bottlenecks: data, engineering, models and gives concrete fixes: prioritize labeled training sets, use transfer learning or unsupervised representation learning from computer vision to cut labels, and build data exchanges to share assets across organizations. Recommends hiring machine learning engineers to run a deployment checklist—model serving, drift detection and scheduled retraining—and points to early work on ML over encrypted data for privacy.
AI-written summary · about
chapters (6)
- 0:00Intro
- 0:18Supervised learning
- 0:31Training data
- 0:52Unsupervised learning
- 1:29Data exchanges
- 2:48Conclusion
related talks
04mManaging Machines
30mChallenges of machine learning: The new ML lifecycle
06mMachine learning is about your data and deployment
41mBeyond the Algorithm: What Makes Machine Learning Work?
27mKeys to Building Machine Learning Systems
26mCompleting the machine learning loop
07mPredictive Modeling and Machine Learning
21mHow The Application Of Machine Learning Is Changing The Way Companies Use Data