Ten interdisciplinary research projects won funding from Princeton University’s Schmidt DataX Fund, aiming to disseminate and deepen the use of artificial intelligence and machine learning on campus to accelerate discovery .
The 10 faculty projects, supported by a major grant from Schmidt Futures, involve 19 researchers and multiple departments and programs, from IT to politics.
Projects explore a variety of topics, including an analysis of how money and politics interact, the discovery and development of new materials exhibiting quantum properties, and the advancement of natural language processing.
“We are excited about the wide range of funded projects, which show the importance and impact of data science across all disciplines,” said Peter Ramadge, Gordon YS Wu professor of engineering at Princeton and director of the Center for Statistics and Machine Learning. (CSML). “These projects use artificial intelligence and machine learning in a variety of ways: to discover connections or hidden models, to model complex systems that are difficult to predict, and to develop new modes of analysis and processing.”
CSML oversees a series of efforts made possible by the Schmidt DataX Fund to extend the reach of data science across campus. These efforts include hiring data scientists and overseeing the awarding of DataX grants. This is DataX’s second round of seed funding, the first being in 2019.
The 10 winning projects and research faculty
Discover development algorithms
Bernard Chazelle, professor of computer science Eugene Higgins; Eszter Posfai, James A. Elkins, Jr. ’41 Preceptor in Molecular Biology and Assistant Professor of Molecular Biology; Stanislav Y. Shvartsman, professor of molecular biology and the Lewis Sigler Institute for Integrative Genomics, and also PhD in 1999. alumnus
“Natural algorithms” is a term used to describe dynamic biological processes built over time through evolution. This project seeks to explore and understand through data analysis a type of natural algorithm, the process of transformation of a fertilized egg into a multicellular organism.
MagNet: Transforming the Concept of Magnetic Power Through Machine Learning
SPICE tools and simulations
Minjie Chen, assistant professor of electrical and computer engineering and the Andlinger Center for Energy and the Environment; Niraj Jha, professor of electrical and computer engineering; Yuxin Chen, Assistant Professor of Electrical and Computer Engineering
Magnetic components are generally the largest and least efficient components in power electronics. To solve these problems, this project proposes the development of an open source magnetic design platform based on machine learning to transform the modeling and design of magnetic power.
Building a multimodal knowledge base for common sense reasoning
Jia Deng and Danqi Chen, assistant professors in computer science
To advance natural language processing, researchers have developed large-scale, text-based common sense knowledge bases that help programs understand facts about the world. But these datasets are laborious to build and have problems with the spatial relationships between objects. This project seeks to address these two limitations by using information from videos with text to automatically build common sense knowledge bases.
Generalized Clustering Algorithms to Map COVID-19 Response Types
Jason Fleischer, Professor of Electrical and Computer Engineering
Clustering algorithms are designed to group objects together, but fail when the objects have multiple labels, the groups require detailed statistics, or the datasets grow or change. This project fills these gaps by developing networks that make clustering algorithms more agile and sophisticated. Improved performance of medical data, in particular patient response to COVID-19, will be demonstrated.
New framework for data in the modeling, characterization and optimization of semiconductor devices suitable for machine learning tools
Claire Gmachl, Eugene Higgins Professor of Electrical Engineering
This project focuses on the development of a new framework based on machine learning to model, characterize and optimize semiconductor devices.
Individual political contributions
Matias Iaryczower, professor of politics
To answer questions about the interplay of money and politics, this project proposes to use micro-level data on the individual characteristics of potential political contributors, the characteristics and choices of political candidates, and political contributions. made.
Building a browser-based data science platform
Jonathan Mayer, Assistant Professor of Computer Science and Public Affairs, Princeton School of Public and International Affairs
Many research problems at the intersection of technology and public policy involve personalized content, social media activity, and other individualized online experiences. This project, which is a collaboration with Mozilla, builds a browser-based data science platform that will allow researchers to study how users interact with online services. The initial study on the platform will analyze how users are exposed to, consume, share and act on political and COVID-19 information and disinformation.
Adaptive depth neural networks and “physical” hidden layers: applications to multiphase flows
Michael Mueller, associate professor of mechanical and aerospace engineering; Sankaran Sundaresan, Professor Norman John Sollenberger in Engineering and Professor of Chemical and Biological Engineering
This project proposes to develop data-based models for complex multi-physical fluid flows using neural networks in which physical constraints are explicitly applied.
Seek to dramatically accelerate the achievement of optimal quantum multibody control using artificial neural networks
Herschel Rabitz, professor of chemistry Charles Phelps Smyth ’16 * 17; Tak-San Ho, research chemist
This project aims to exploit artificial neural networks to design, model, understand and control quantum dynamics phenomena between different particles, such as atoms and molecules. (Note: This project also received DataX funding in 2019.)
Discovery and design of the next generation of topological materials using machine learning
Leslie Schoop, assistant professor of chemistry; Bogdan Bernevig, professor of physics; Nicolas Regnault, visiting researcher in physics
This project aims to use machine learning techniques to discover and develop “topological matter”, a type of matter that exhibits quantum properties, whose future applications may have an impact on energy efficiency and the rise of computers. super quantum. Current applications of the topological material are severely limited because its desired properties only appear at extremely low temperatures or at high magnetic fields.