Algorithmic, Mathematical, and Statistical Foundations of Data Science and Applications

April 12-13, 2019

About

Data Science is a growing field that uses data and computing to improve everyday life. This field poses a unique set of multi-disciplinary challenges spanning:

  • Computer science
  • Statistics
  • Applied mathematics
  • Machine learning

and their use in the sciences (such as biology, chemistry, physics, engineering, sociology, economics, medicine, etc.). The data science workshop at Purdue will focus on the theoretical foundations of Data Science while highlighting the helpful feedback cycle between foundational work and applications.

Program

Friday, April 12

(at the Shively Club, 3rd floor of Ross-Ade Pavilion,

850 Steven Beering Dr, West Lafayette, IN 47906)

Time Event
8:30am - 9:15am Registration & Coffee
9:15am - 9:30am Introductory Remarks
9:30am - 10:45am

Session 1

On the Theory of Gradient-Based Learning: A View from Continuous Time - Abstract
Michael Jordan
9:30 - 10:15am

Controlling Confounding and Selection Biases in Causal Inference - Abstract
Elias Bareinboim
10:15 - 10:30am

Engineering Drug Discovery using Chemical Data Science - Abstract
Gaurav Chopra

10:30 - 10:45am

10:45am - 11:15am Coffee Break
11:15am - 12:30pm

Session 2

Stochastic Gradient Descent, in Theory and Practice - Abstract
Rachel Ward
11:15am - 12:00pm

Learning to Grow Economies - Abstract
Simina Branzei
12:00pm - 12:15pm

From ADMM to Consensus Equilibrium - Abstract
Stanley Chan
12:15pm - 12:30pm

12:30pm - 2:00pm Lunch Break
2:00pm - 3:15pm

Session 3

Graph Mining at Scale: From Theory to Practice and Back - Abstract
Vahab Mirrokni

2:00pm - 2:45pm

Statistical methods for prediction and anomaly detection in dynamic networks - Abstract
Jennifer Neville

2:45pm - 3:00pm

Leveraging Big Data to Understand the Genetics of Health and Disease - Abstract
Peristera Paschou
3:00pm - 3:15pm

3:15pm - 3:45pm Coffee Break
3:45pm - 5:00pm

Session 4

Computation in the Brain - Abstract
Christos Papadimitriou
3:45pm - 4:30pm

Information Content in Dynamic Networks - Abstract
Wojciech Szpankowski
4:30pm - 4:45pm

Brain Connectomics: From Maximizing Subjects Identifiability to Disentangling Heritable and Environment Traits - Abstract
Joaquin Goni
4:45pm - 5:00pm

 

Saturday, April 13

(at Lawson Computer Science Building (LWSN), Room 1142)

Time Event
9:00am - 9:30am Breakfast 
9:30am - 10:45am

Session 5

Stochastic Optimization for Large-Scale Tensor Decomposition - Abstract
Tammy Kolda
9:30 - 10:15am

Stein Goodness-of-fit Tests for Discrete and Point Process Data - Abstract
Vinayak Rao
10:15 - 10:30am

Data Science Case Studies at Purdue and Sparse Bayesian Deep Learning - Abstract
Guang Lin
10:30 - 10:45am

10:45am - 11:15am Coffee Break
11:15am - 12:15pm

Session 6

Sparse Matrices in Sparse Analysis - Abstract
Anna Gilbert
11:15am - 12:00pm

Goodness of Fit Testing for Dynamic Network Models - Abstract
Abram Magner
12:00pm - 12:15pm

12:15pm - 12:30pm

Closing Remarks

Speakers

Keynote Speakers

Anna Gilbert

Sparse Matrices in Sparse Analysis

University of Michigan

Department of Mathematics

Michael Jordan

On the Theory of Gradient-Based Learning: A View from Continuous Time

University of California Berkeley

Department of Electrical Engineering and Computer Sciences, Department of Statistics

Tamara Kolda

Stochastic Optimization for Large-Scale Tensor Decomposition

Sandia National Labs

Vahab Mirrokni

Google Research

Christos Papadimitriou

Computation in the Brain

Columbia University

Department of Computer Science

Rachel Ward

Stochastic Gradient Descent, in Theory and Practice

University of Texas at Austin

Department of Mathematics

Purdue Highlight Speakers

Elias Bareinboim

Controlling Confounding and Selection Biases in Causal Inference

Department of Computer Science

Ilias Bilionis

Learning from Small Data by Exploiting Physics

Department of Mechanical Engineering

Simina Branzei

Learning to Grow Economies

Department of Computer Science

Stanley Chan

From ADMM to Consensus Equilibrium

Department of Electrical and Computer Engineering, Statistics

Gaurav Chopra

Department of Chemistry

Joaquin Goni

Brain Connectomics: From Maximizing Subjects Identifiability to Disentangling Heritable and Environment Traits

Department of Industrial Engineering

Guang Lin

Data Science Case Studies at Purdue and Sparse Bayesian Deep Learning

Department of Mathematics

Abram Magner

Goodness of Fit Testing for Dynamic Network Models

University of Michigan

Jennifer Neville

Department of Computer Science

Peristera Paschou

Leveraging Big Data to Understand the Genetics of Health and Disease

Department of Biological Sciences

Vinayak Rao

Stein Goodness-of-fit Tests for Discrete and Point Process Data

Department of Statistics

Wojciech Szpankowski

Information Content in Dynamic Networks

Department of Computer Science

Registration

Click here to register. Registration is complimentary for (Purdue or non-Purdue) undergraduates; $10 for Purdue graduate students; $20 for Purdue faculty, Purdue postdocs, and non-Purdue graduate students; $40 for non-Purdue faculty or non-Purdue postdocs; and $60 for non-academic participants.

Logistics

The conference will be held in two different locations on Friday and Saturday. 

Friday

We will be in the Shively Club, 3rd floor of Ross-Ade Pavilion at the Purdue Stadium. The address is 850 Steven Beering Dr, West Lafayette, IN 47906. For parking, please feel free to park in the North Stadium Lot, immediately north of the stadium. The signs say parking for ABC permits only, but we have negociated an arrangement for people attending our conference. Please feel free just to park there. 

Map to Parking at Ross-Ade

Saturday

We will be in the Lawson Computer Science Building 305 N. University St. West Lafayette, IN 47907, room 1142. Parking is available in the adjacent parking garage (parking garage entrance just south of Third Street on University Street). 

Parking map to Lawson

Organizers

Petros Drineas

Department of Computer Science

Purdue University

David Gleich

Department of Computer Science

Purdue University