AutoML—hype vs. value
summary
Argues that AutoML rarely replaces experts but speeds work by automating modeling tasks; Oracle demonstrates OML4PY functions for automated algorithm selection, feature selection and gradient‑descent hyperparameter tuning, plus a no‑code AutoML UI that runs an experiment on a wine dataset where an SVM with Gaussian kernel is ranked best, produces a tuned model, generates an editable OML4PY notebook and publishes the model to an OML Services REST endpoint for immediate deployment.
AI-written summary · about
chapters (8)
- 0:00Introduction to AutoML
- 1:16Demystifying AutoML
- 4:56The machine learning process
- 10:20Automation feasibility
- 14:28The modeling time sink
- 17:07Value and business benefits
- 21:03Oracle AutoML components
- 24:31Demo and final thoughts
related talks
31mIn-database machine learning: Automate, scale, deploy
39mAuto ML
33mML: Automated Machine Learning
08mBroadening AI Adoption with AutoML
21mDemocratizing AI: AutoML Transforms Predictive Analytics
22mPredict the future (or don't) with machine learning in applications
38mFantastic ML apps and how to build them
01h 08mAutoML: 40 hours of doing machine learning without knowing it