Symposium 2017

The 2017 MINDS Symposium was the inaugural event of the Johns Hopkins Mathematical Institute for Data Science, held November 3, 2017 in Baltimore.

The agenda for the 2017 MINDS Symposium featured research from experts at the world’s top academic centers for data science.



8:15 a.m. Check-In and Registration

Opening Ceremony

9:00 a.m. Dean’s Welcome, Ed Schlesinger, Johns Hopkins Whiting School of Engineering
9:05 a.m. Vice Provost’s Welcome, Denis Wirtz, Johns Hopkins University
9:10 a.m. Director’s Inaugural Address and Welcome, René Vidal, Johns Hopkins Mathematical Institute for Data Science

Session 1 | Data Science Foundations and Applications in Medicine

9:25 a.m. On Computational Thinking, Inferential Thinking, and Data ScienceMichael I. Jordan, University of California, Berkeley
10:05 a.m. The Role of Data in Achieving Precision and Value in HealthcareGregory D. Hager, Johns Hopkins Malone Center for Engineering in Healthcare
10:20 a.m. Machine Learning Approaches for Personal GenomicsAlexis Battle, Johns Hopkins University


10:35 a.m. Coffee Break

Session 2 | Reinforcement Learning, Representation Learning, and Applications in Vision and Speech

11:00 a.m. Conditional Mean Embeddings for Reinforcement LearningJohn Shawe-Taylor, University College London
11:40 a.m. Designing and Learning Representations for Visual Data in the Age of Deep LearningStefano Soatto, Amazon Web Services and University of California, Los Angeles
12:20 p.m. Alexa, Tell Me How Kaldi and Deep Learning Revolutionized Automatic Speech Recognition!Sanjeev Khudanpur, Johns Hopkins Center for Language and Speech Processing


12:35 p.m. Lunch Break

Session 3 | Random Matrices, Interactive Machine Learning, and Applications in Astronomy

2:00 p.m. Applied Random Matrix TheoryJoel A. Tropp, California Institute of Technology
2:40 p.m. Toward an Era of Intelligent Interactive AlgorithmsAarti Singh, Carnegie Mellon University
3:20 p.m. Data Science in AstronomyAlex Szalay, Johns Hopkins Institute for Data Intensive Engineering and Science


3:35 p.m. Coffee Break

Session 4 | Non-Convex Optimization, Manifold Learning, and Applications in Computational Anatomy

4:00 p.m. Non-Convex Optimization for Low-Complexity Signal and Data ModelingJohn Wright, Columbia University
4:40 p.m. Is Manifold Learning for Toy Data Only?Marina Meila, University of Washington
5:20 p.m. Theoretical and Numerical Challenges in Medical Image Analysis and Computational AnatomyNicolas Charon, Johns Hopkins Center for Imaging Science


5:35 p.m. Closing Remarks