Automated ML: A journey from CRISPR.ML to Azure ML
summary
A meta‑learning system turns CRISPR.ML’s months of manual tuning into a recommender over a discrete space of preprocessing, model families and hyperparameters, using a probabilistic exploration/exploitation model to pick pipelines and reason across datasets. It produces a ready Python object, integrates into Azure ML and Power BI, and ships with model explainability; applied to gene‑editing data it reports 20% improvement in accuracy and 50% savings in time per gene while outputting a deployable Python object.
AI-written summary · about
chapters (5)
- 0:00Start
- 0:40Building a Model
- 2:05The data scientist's dilemma...
- 3:44Machine learning predictive modelling for CRISPR
- 10:04Microsoft Al breakthroughs
related talks
58mDeep Dive into Azure Automated Machine Learning
49mAutomated Machine Learning on Azure
41mLeveraging Automated Machine Learning to enable anyone to develop Machine Learning solutions
51mTaking Models to the Next Level with Azure Machine Learning
24mFrom Model to Production: Automating the ML Lifecycle with CI/CD
48mMachine Learning in Azure Databricks
34mImproving the Fairness and Reliability of AI Solutions
01h 14mAI of Things