ICME Xpo Research Symposium

ICME's annual research symposium, Xpo, was held virtually on Thursday, May 28, 2020. This event was held for Stanford faculty, staff, students, and external partners of ICME.

Every May, ICME holds a Research Symposium, providing an up-close and inside look at current research and future plans for ICME faculty and students. ICME is engaged with over 50 faculty from 20-plus departments throughout Stanford. This is a unique opportunity to see how computational mathematics, data science, machine learning, scientific computing, and related fields are applied across a wide range of domain areas.

Speaker Bios Research Presentations Agenda

Computational Math is Everywhere

Presented by Prof. Gianluca Iaccarino, ICME Director

• Update on ICME research, curriculum, and impact

Computational Math: Human Centered

Moderated by Alison Marsden, Prof. of Pediatrics and Bioengineering

• Gabe Maher, ICME PhD Student - "Acceleration and Geometric Uncertainty Quantification of Patient-Specific Cardiovascular Modeling with Neural Networks"
• Julia Palacios, Prof. of Statistics and of Biomedical Data Science - "Phylodynamics of Covid-19"
• Sylvia Plevritis, Prof. and Chair of Biomedical Data Science and of Prof. of Radiology - "Computational Modeling of Tumor Microenvironment Interactions"
• Q&A

Computational Math: Society Centered

Moderated by Ashwin Rao, ICME Adjunct Faculty 

• Ashish Goel, Prof. of Management Science & Engineering - "Decision Making at Scale"
• Allison Koenecke, ICME PhD student - "Racial Disparities in Automated Speech Recognition"
• Alan Gous, ICME Faculty - "What’s the use of privacy?"
• Peter Glynn, Prof. of Management Science & Engineering - "Analytics When the Data Quality is Bad: COVID-19 Hospital Operations"
• Q&A

Computational Math: Planet Centered

Moderated by Prof. Gianluca Iaccarino

• Catherine Gorlé, Prof. of Civil and Environmental Engineering - "Computational Wind Engineering for Sustainable Buildings and Cities"
• Rahul Sarkar, ICME PhD student - "Can a Quantum Computer Help in Creating a Sustainable Future?"
• Biondo Biondi, Prof. of Geophysics - "Taking the Pulse of our Cities by Fiber-Optic Seismology"
• Q&A

AI for Good

Karen Matthys, ICME Executive Director, External Partners and AI for Good Co-Instructor

• Key takeaways from the ICME 2020 Seminar Class

NeurIPS 2019 Highlights

• James Zou, Prof. of Biomedical Data Science - "How to Make AI Forget You? Adventures in Making Algorithms More Grown-Up."
• ICME PhD Student Panel: Tetiana Parshakova, Junzi Zhang, and Ines Chami
• Q&A

 (Video begins at 14:32)

NeurIPS 2019 Highlights

A number of ICME graduate students and affiliated faculty attended and gave presentations at the Neural Information Processing Systems (NeurIPS) 2019 Conference. Here are some of their papers (listed alphabetically by title):

• Almost Horizon-Free Structure-Aware Best Policy identification with a Generative Model - Andrea Zanette, Mykel J. Kochenderfer & Emma Brunskill
• Distributional Reinforcement Learning for Energy-Based Sequential Models - Tetiana Parshakova, Jean-Marc Andreoli & Marc Dymetman
• Hyperbolic Graph Convolutional Neural Networks - Ines Chami, Zhitao Ying, Christopher Ré & Jure Leskovec
• Learning Mean-Field Games- Xin Guo, Anran Hu, Renyuan Xu, Junzi Zhang
• Limiting Extrapolation in Linear Approximate Value Iteration - Andrea Zanette, Alessandro Lazaric, Mykel J. Kochenderfer & Emma Brunskill
• Low-Dimensional Knowledge Graph Embeddings via Hyperbolic Rotations - Ines Chami, Adva Wolf, Frederic Sala & Christopher Ré
• Making AI Forget You: Data Deletion in Machine Learning- James Zou, Antonio Ginart, Melody Guan, Gregory Valiant
• Multivariate Distributionally Robust Convex Regression under Absolute Error Loss - Jose Blanchet, Peter Glynn, Jun Yan, & Zhengqing Zhou
• Multiview Aggregation for Learning Category-Specific Shape Reconstruction -Srinath Sridhar, Davis Rempe, Leonidas J Guibas
• Towards Automatic Concept-based Explanations - James Zou, Been Kim, Amirata Ghorbani, James Wexler
• You Only Propagate Once: Accelerating Adversarial Training via Maximal Principle- Dinghuai Zhang, Tianyuan Zhang, Yiping Lu, Zhanxing Zhu, Bin Dong