1. (Full-day at Salon A) Intermediate Machine Learning: Key Concepts and Techniques

Instructor

Dr. David Rosenberg is a data scientist in the data science group in the Office of the CTO at Bloomberg, and an adjunct associate professor at the Center for Data Science at New York University, where he has repeatedly received NYU’s Center for Data Science ``Professor of the Year” award. He received his Ph.D. in statistics from UC Berkeley, where he worked on statistical learning theory and natural language processing. David received a Master of Science in applied mathematics, with a focus on computer science, from Harvard University, and a Bachelor of Science in mathematics from Yale University.

Outline

There is no shortage of tutorials describing a huge variety of machine learning models and techniques. In this short course, we take a step back and present a coherent framework for thinking about supervised machine learning models in general. We examine multiple examples of the four fundamental components of a machine learning method: loss function, regularization, hypothesis space, and optimization method. Within this framework, we'll study linear regression (regression, lasso, and elastic net) and classification methods. We'll also introduce the most important nonlinear models, including tree-based ensemble methods and neural network models. Time-permitting, we may discuss conditional probability models, noting that the vast majority of contemporary deep learning models are of this type, as well as an approach to multiclass classification that generalizes to structured prediction and ranking problems, among many other applications. Throughout our discussion, we will introduce the terminology and notation used by experts in machine learning to help bridge the gap between introductory-level tutorials and the more advanced materials you can find at conferences and in graduate-level courses.

Prerequisites

Familiarity with basic mathematical notation (e.g., ∑ for summation, arg min ), basic linear algebra (e.g., matrix multiplication, projections, inner products, norms, and spans), and introductory probability (probability distributions, conditional probability and conditional expectation).

2. (Full-day at Salon B) Big Data Analytics and Deep Learning

Instructors

Dr. Hui Lin is leading and building data science department at Netlify. Before Netlify, she was a Data Scientist at DuPont. She was a leader in the company of applying advanced data science to enhance Marketing and Sales Effectiveness. She provided data science leadership for a broad range of predictive analytics and market research analysis from 2013 to 2018. She is the co-founder of Central Iowa R User Group, blogger of scientistcafe.com and 2018 Program Chair of ASA Statistics in Marketing Section. She enjoys making analytics accessible to a broad audience and teaches tutorials and workshops for practitioners on data science. She holds MS and Ph.D in statistics from Iowa State University.

Dr. Ming Li is currently a Research Scientist at Amazon. He organized and presented 2018 JSM Introductory Overview Lecture: Leading Data Science: Talent, Strategy, and Impact. He was the Chair of Quality & Productivity Section of ASA for 2017. He was a Data Scientist at Walmart and a Statistical Leader at General Electric Global Research Center. He obtained his Ph.D. in Statistics from Iowa State University in 2010. With deep statistics background and a few years’ experience in data science, he has trained and mentored numerous junior data scientist with different background such as statistician, programmer, software developer, database administrator and business analyst. He is also an Instructor of Amazon’s internal Machine Learning University and was one of the key founding member of Walmart’s Analytics Rotational Program which bridges the skill gaps between new hires and productive data scientists.

Outline

In the past couple of years, deep learning has gained traction in many areas. It becomes an essential tool in data scientist’s toolbox. In this course, students will develop a clear understanding of the big data cloud platform, technical skills in data sciences and machine learning, the motivation and use cases of deep learning through hands- on exercises. We will also cover the “art” part of data science: data science project flow, general pitfalls in data science and machine learning, and soft skills to effectively communicate with business stakeholders. The course is for audience with statistics background. We use real-world data science and machine learning problems to illustrate data science workflow, pitfalls, and soft skills. The hands-on sessions use Databricks community edition cloud platform. Specific modules are: (1) Big data platform using Spark through R sparklyr package; (2) Introduction to deep neural network, convolutional neural network recurrent neural networks, and their applications; (3) Deep learning examples using TensorFlow through R keras package.

Prerequisites

Introductory statistics or practical experience in data science ; entry level of R knowledge; a free Databrick Community Edition account through https://databricks.com/try-databricks; a laptop.