Beyond P-values Machine Learning

04/17/2018 - 8:00am to 04/19/2018 - 5:00pm
ML - Damon Room

 

beyond p bear

Image: Courtesy of Philippe Naveau

Course Description

This three-day course, April 17-19, 2018, will provide an introduction to fundamental methods used in Machine Learning.  We will start with dimension reduction methods, which are often used as precursors to subsequent analysis.  This will be followed by an overview of unsupervised vs. supervised learning.  For unsupervised learning, we will cover various cluster analysis methods such as k-means.  For supervised learning, we will introduce data-driven approaches such as regression trees and modeling-based approaches with a special focus on artificial neural networks and deep learning.  The course is aimed at an applied audience and will make heavy use of data examples to illustrate the concepts.  We'll use the open-source statistical software R [https://www.r-project.org/]. The format of the course is hands-on and participants will use their own laptops.

Instructors

The lead instructor for the course is Valerie Monbet, Professor of Statistics at the University of Rennes . She will be assisted by graduate students and post-doctoral fellows specializing in Statistics and Machine Learning. Seats are limited to 12 participants to allow for effective one-on-one coaching. To apply, please visit the Machine Learning Application link on the left hand-side of the workshop webpage.   Note the application deadline is March 2, at 5:00 PM MST.

This training is for UCAR employees only.  For more information, see the webpage here.  The application window has closed.

 

Agenda:  Beyond P-values Course: Machine Learning

 

Tuesday, April 17, 2018, NCAR Mesa Lab - Damon Room (239)

9:00am – 10:15am

Module 1: Introduction + Dimension Reduction

10:15am – 10:45am

Break

10:45am – 12:00pm

Module 2:  Clustering

12:00pm – 1:00pm

Lunch

1:00pm – 2:15pm

Module 3:  Linear Models (I): optimization and estimation, model validation

2:15pm – 2:45pm

Break

2:45pm – 4:00pm

Module 4:  Linear Models (II): variable selection

Wednesday, April 18, 2018, NCAR ML- Damon Room (239)

9:00am – 10:15am

Module 5:  Data driven methods/nearest neighbors/curse of dimensionality

10:15am – 10:45am

Break

10:45am – 12:00pm

Module 6:  Decision trees and variable importance (+MARS algorithm)

12:00pm – 1:00pm

Lunch

1:00pm – 2:15pm

Module 7:  Bagging and random forest

2:15pm – 2:45pm

Break

2:45pm – 4:00pm

Module 8:  Boosting & gradient boosting

Thursday, April19, 2018, NCAR ML - Damon Room (239)

9:00am – 10:15am

Module 9: Neural Networks (I)

10:15am – 10:45am

Break

10:45am – 12:00pm

Module 10:  Neural Networks (II)

12:00pm –  1:00pm

Lunch

1:00pm – 2:15pm

Module 11:  Deep Learning – Weather Applications

2:15pm –  2:45pm

Break

2:45pm – 4:00pm

Module 12:  Deep Learning