Azure Applied AI and AutoML

This is the presentation I used for a session at Microsoft Turkey Startup Bootcamp where we discussed Azure Applied AI Services and the use of AutoML.

Transcript

1. Azure Applied AI and AutoML
   Daron Yöndem
   Azure Application Development Lead for MEA
   Microsoft
   http://daron.me
   @daronyondem
   

   View Slide

2. Different flavors of AI/ML
   Control Productivity
   Machine
   Learning
   Supervised, Unsupervised,
   and reinforcement learning
   AutoML
   Automation for iterative tasks of ML
   AI
   Understand \ Interpret\ Learn\ and
   make decisions.
   Applied AI
   task-specific AI, and business
   logic as turnkey AI services.
   Azure ML Cognitive Services
   ML AI
   

   View Slide

3. Azure AI
   

   View Slide

4. What is Applied AI?
   AI models
   

   View Slide

5. What are Azure Applied AI Services?
   Extract actionable
   insights from your
   videos
   Proactively monitor
   metrics and diagnose
   issues
   Bring AI-powered cloud search
   to your mobile and web apps.
   Extract actionable
   insights from your
   videos
   Turn documents into
   usable data at a fraction of
   the time and cost
   Help users read and
   comprehend text
   

   View Slide

6. No Code Chatbot
   Demo
   

   View Slide

7. What is AutoML?
   • Automated Machine Learning (AutoML) is the process of
   automating end-to-end process of applying Machine Learning
   (ML) to create, develop and deploy predictive models so that
   any enterprise benefits from data.
   • Low-code / No-Code Experience
   • Keeps searching for the algorithm and hypermeters based on
   the metrics you define.
   

   View Slide

8. The Process
   SQL DB
   Cosmos DB
   Datawarehouse
   Data lake
   Blob storage
   …
   Prepare Data Build & Train Deploy
   

   View Slide

9. Sample Problem
   How much is this car worth?
   

   View Slide

10. Model creation takes time… a lot o time…
    Which algorithm? Which parameters?
    Which features?
    Mileage
    Condition
    Car brand
    Year of make
    Regulations
    …
    Gradient Boosted
    Nearest Neighbors
    SVM
    Bayesian Regression
    LGBM
    …
    Nearest Neighbors
    50%
    Model
    Iterate
    30%
    Gradient Boosted
    Mileage
    Car brand
    Year of make
    Car brand
    Year of make
    Condition
    Parameter 1
    Parameter 2
    Parameter 3
    Parameter 4
    …
    

    View Slide

11. The Complexity
    

    View Slide

12. Model Selection & Hyperparameter Tuning
    Dataset
    Training
    Algorithm 1
    Hyperparameter
    Values – config 1
    Model 1
    Hyperparameter
    Values – config 2
    Model 2
    Hyperparameter
    Values – config 3
    Model 3
    Model Training
    Infrastructure
    Training
    Algorithm 2
    Hyperparameter
    Values – config 4
    Model 4
    Repetitive & Manual
    

    View Slide

13. Introducing Automated Machine Learning
    Dataset
    Optimization
    Metric
    Constraints
    (Time/Cost)
    ML Model
    Automated ML
    Accessible & Faster
    

    View Slide

14. How to use AutoML in Azure?
    • Code: Azure Machine Learning Pyhton SDK
    • No-Code: Azure Machine Learning Studio
    

    View Slide

15. Different tasks
    • Classification: Fraud Detection, Marketing Prediction
    • Regression: Predict pricing
    • Time-series forecasting: Sales and Demand Forecasting
    

    View Slide

16. Data Scientist in a Box
    Demo
    

    View Slide

17. Thanks
    http://daron.me | @daronyondem
    Grab slides on http://decks.daron.me/
    

    View Slide