Machine Learning 101

Description

TUTORIAL PREREQUISITES

Welcome to Machine Learning 101! We will be learning the basics of machine learning by building real applications: recommender and image analysis with deep learning.

The tutorial landing page (https://dato.com/events/training/2015-strata-nyc.html) contains the agenda for the day as well as relevant material. Please follow the instructions for setting up your laptop PRIOR to arriving onsite.

If you run into any problems during installation, please email training@dato.com.

TUTORIAL DESCRIPTION

This hands-on tutorial provides a quick start to building intelligent business applications using machine learning. Learn about machine learning basics, feature engineering, recommender systems, and deep learning. We will build and deploy large-scale machine learning applications with Dato’s Machine Learning platform: GraphLab Create, Dato Distributed, and Dato Predictive Services. The program will center around building two applications: a content-based recommender that tells you which talks you might be interested in at Strata, and an image search application built using deep learning.

We will walk you through all the steps of prototyping and production: data cleaning, feature engineering, model building and evaluation, and deployment.

Please check back here prior to the tutorial date for installation instructions.

Topics:

• Overview of machine learning
• Feature engineering
• Personalized recommenders
• Content-based analysis
• Factorization models
• Building StrataNow—a personalized talk recommender for StrataConf
• Deep learning
• Deep learning models
• Transfer learning and deep features
• Building an image search application to find similar clothing
• Deploying machine learning models in production
• Constructing a real-time predictive service
• Monitoring and evaluation