Broadening AI Adoption with AutoML
summary
AutoML replaces repetitive pipeline steps with automated feature engineering, automated model selection and hyperparameter tuning, letting domain experts produce a solid baseline without ML expertise. Demonstration trains an activity classifier from accelerometer streams, explores tradeoffs for mobile and embedded targets including memory, power and model size, and shows the automated pipeline producing a deployable model plus where manual optimization still improves accuracy. Tools used are AutoML components integrated into the workflow.
AI-written summary · about
related talks
39mAuto ML
01h 08mAutoML: 40 hours of doing machine learning without knowing it
21mDemocratizing AI: AutoML Transforms Predictive Analytics
33mML: Automated Machine Learning
47mBuilding smarter solutions without machine learning expertise
56mBuilding Intelligent Apps with Machine Learning
28mHow to boost data science routines with DataArt’s AutoML Platform?
01h 14mAI of Things