Machine Learning Explained

Actions Panel

Machine Learning Explained

When and where

Date and time


Quinlan School of Business 16 East Pearson Street Schreiber Center Room 302 Chicago, IL 60611

Map and directions

How to get there


Machine learning, data science, data mining, predictive analytics... all are hot areas in research and industry. This talk is aimed at explaining the common underlying principles across these domains; this will provide context for explaining state of the art techniques that are creating buzz for good reason - from the universal benefits of Big Data to deep learning and ensemble learning methods. The talk begins but noting the common pitfalls in learning from data, drawing analogies from human learning. This will progress to an explanation of the "standard form" for learning from data (for the initiated: samples x features tables and targets) and how you can frame literally any learning problem in that form (with varying levels of success, but literally ANY learning problem).

Although there are numerous classifiers and regression models, there are common features to all of them that can permit a user to simply plug-and-play models, in many cases by only changing one line of code. The nearly endless supply of models and modifications may be intimidating at first, but we will discuss the wonderful way that data can be leveraged to definitively answer "which is the best approach?". The talk will be followed up with links to - and with time permitting discussion of - tutorials where you can immediately see these principles in practice and try them for yourself.


4:00pm - 4:30pm

Appetizers and Soft Drinks

4:30pm - 6:00pm



Ting Xiao, Ph.D.


Ting Xiao is a postdoctoral researcher in particle physics at Northwestern University. She uses statistical and computational tools to extract meaningful signals from large data sets. She has been applying these skills on data collected from particle accelerators at Cornell and Jefferson National Lab, where she has improved the precision of various physical measures, allowing for the accurate detection of a variety of subatomic particles. She received her PhD from Northwestern University in 2016 and her undergraduate degree at Zhejiang University in China.

Her research interests are in data science, with a focus on signal processing in a variety of domains; this includes multiple current projects with Loyola students using probabilistic graphical models for statistical inference on auditory signals, video, and inertial sensor data. She is a coauthor on 60 papers working with a variety of collaborations, amassing over 1000 citations; the most cited paper among these is a first author paper in which she discovered a new particle - Zc0(3900). She is also currently teaching Loyola's first course COMP 180: Computing and Data Analysis for the Sciences.