San Francisco, CA

October 15-17, 2017

101 Howard Street | view map >


Introduction to Deep Learning with Python
Jeremy Howard
Co-Founder, Former President Kaggle, Distinguished Scholar in Deep Learning USF Data Institute
Yannet Interian
Assistant Professor Analytics, USF

Deep learning is the state-of-the-art machine learning technique in areas such as object recognition, image segmentation, speech recognition and machine translation. This workshop introduces you to the basic concepts of deep learning. It covers practical details of real deep learning applications. You will learn hands-on model building using Python, Keras and TensorFlow.

For best use of the workshop, you should have some experience with numpy. Make sure you are comfortable with the techniques shown in this tutorial.

Designed Experiments for Data Scientists
Nathaniel Stevens
Assistant Professor Statistics, USF

Much of the buzz surrounding data science has been associated with machine learning and its successes in finding patterns and relationships in data and then leveraging these correlations for purposes of prediction and classification. However when causal inference is required, a carefully designed experiment is necessary to evaluate the impact of altering one or more variables on some outcome of interest. In this workshop we explore available methods for AB testing and experimentation to identify and quantify cause-and-effect relationships for the purpose of optimization.

Network Analytics
James D. Wilson
Assistant Professor Statistics, USF

Complex networks have generated new and exciting areas of analysis, particularly in statistics and computation. These inherently high dimensional data objects present unique challenges in modeling, simulation, and analysis that require innovative computational techniques. Networks provide important information about data with relational structure, and often illuminate insights beyond multivariate analysis alone. In this workshop, we will discuss three fundamental aspects of network data: exploratory analysis, statistical models, and community detection. We will survey common network analysis techniques from machine learning and random graph theory and discuss how to implement these techniques in the R programming language. We will investigate several industrial case studies that highlight the importance of network analysis in practice.

How to Hire a Data Scientist
Jason Shu
SVP Data Science at Aki Technologies

Historically, technical roles have been filled by specialists trained to solve specific technical problems. Data Science is unique in that it typically requires proficiency in many different areas, but perhaps expertise in none. This jack-of-all-trades position results in companies testing a variety of rote skills to determine diversity of candidates’ knowledge, but does this process identify the skills that are truly needed to help an organization?

In this workshop, we will compare the explicit and implicit needs of companies’ data science teams and discuss how this distinction can alter how candidates could be evaluated. This workshop is intended for hiring managers looking to consider outside-the-box approaches to interviewing, but it would also benefit aspiring data scientists who want to think about the field in ways beyond just algorithms and formulas.