Events

AI and Future of Healthcare

AI technology is permeating every aspect of healthcare. It is transforming how care is provided in acute, ambulatory and home settings. It is transforming how population health is being managed. AI in combination with other disciplines, such as genetics and virtual reality, is accelerating innovative treatments. Drug discovery and clinical trials are being transformed. This presentation will provide an overview of where AI is already being used and prognosis for the immediate future.

About V. “Juggy” Jagannathan, PhD
V. “Juggy” Jagannathan, PhD, researches artificial intelligence and computer science for 3M Health Information Systems, developing new natural language and deep learning technologies that automatically structure patient-physician conversations into clinical documents. As an AI researcher with over four decades in the field, Juggy’s background includes both industry technology development and academia. The company he founded in 1996 with technology from West Virginia University (WVU) was acquired by M*Modal in 2003 and he has been with the company ever since. He is also an adjunct professor teaching Computer Science and Natural Language Processing (NLP) at WVU. Juggy is a tennis nut, golf nut and a yoga nut. In short, a nut.

The Role of Social Networks in Economic Mobility

Social capital—the strength of an individual’s social network and community—has been identified as a potential determinant of outcomes ranging from education to health. Mike Bailey and his team used data on 21 billion friendships in the U.S. to measure and analyze different types of social capital including connectedness between different types of people, social cohesion, and civic engagement. The researchers demonstrate the importance of distinguishing these forms of social capital by analyzing their associations with economic mobility across areas. The share of high-SES friends among individuals with low SES—which the researchers term economic connectedness—is among the strongest predictors of upward income mobility identified to date.

In a different paper, Bailey and his team used social network data in India to show the importance of social networks to labor migrants and find that increasing social connectedness across space may have considerable economic gains, improving average wages by 3 percent (24 percent for the bottom wage-quartile) in a migration model.

Papers:

• Social capital I: measurement and associations with economic mobility
• Social capital II: determinants of economic connectedness
• Social Networks and Spatial Mobility: Evidence from Facebook in India

About Mike Bailey
Mike Bailey is a senior social scientist at Meta on the Core Data Science team. His work focuses on the role of social networks on economic opportunity including migration, health, education, and social capital and has been featured in top scientific journals such as Nature and the Journal of Political Economy and covered by outlets such as The Economist and The New York Times.

Leveraging machine learning and omics toolsets to elucidate genomic determinants of nanoparticle delivery

Clinical translation of nanomedicine is hampered by limited accumulation at target disease sites. The complexity and heterogeneity of both biological environment and nanoparticle construct make it prohibitively challenging to deconvolute individual factors that drive nanocarrier targeting and accumulation. We have leveraged library-based and pooled cell screening approaches to gain a holistic understanding of both nanomaterials properties and biological features that mediate successful drug delivery. Through integration of omics data via supervised machine learning, we have identified predictive genomic features that modulate uptake of specific nanocarriers. In this talk, I will describe our nanoparticle library and pooled screen design as well as the results from these integrated screens, which indicate that matching nanocarrier properties and genomic profiles may be a key component to achieving targeted delivery.
 
About Natalie Boehnke
Natalie Boehnke is a new assistant professor in the Department of Chemical Engineering and Materials Science at the University of Minnesota. She received her Ph.D. in chemistry from UCLA in 2017 and recently completed postdoctoral training at the Koch Institute for Integrative Cancer Research at MIT. Her research interests include gaining a fundamental understanding of the mechanisms mediating drug delivery to accelerate clinical translation of nanomedicine, focusing on using high throughput screening, omics, and machine learning approaches.

The power of adaptivity in representation learning

From meta-learning to federated learning

A central problem in machine learning is as follows: How should we train models using data generated from a collection of clients/environments, if we know that these models will be deployed in a new and unseen environment?

In the setting of few-shot learning, two prominent approaches are: (a) develop a modeling framework that is “primed” to adapt, such as Model Adaptive Meta Learning (MAML), or (b) develop a common model using federated learning (such as FedAvg), and then fine tune the model for the deployment environment. We study both these approaches in the multi-task linear representation setting. We show that the reason behind generalizability of the models in new environments trained through either of these approaches is that the dynamics of training induces the models to evolve toward the common data representation among the clients’ tasks.

In both cases, the structure of the bi-level update at each iteration (an inner and outer update with MAML, and a local and global update with FedAvg) holds the key — the diversity among client data distributions are exploited via inner/local updates, and induces the outer/global updates to bring the representation closer to the ground-truth. In both these settings, these are the first results that formally show representation learning, and derive exponentially fast convergence to the ground-truth representation. Based on joint work with Liam Collins, Hamed Hassani, Sewoong Oh, and Sanjay Shakkottai.

Papers: 

• MAML and ANIL Provably Learn Representations

• FedAvg with Fine Tuning: Local Updates Lead to Representation Learning

About Aryan Mokhtari
Aryan Mokhtari is an Assistant Professor in the Electrical and Computer Engineering Department of the University of Texas at Austin (UT Austin) where he holds the Fellow of Texas Instruments/Kilby. Before joining UT Austin, he was a Postdoctoral Associate in the Laboratory for Information and Decision Systems (LIDS) at MIT.  Prior to that, he was a Research Fellow at the Simons Institute for the program on “Bridging Continuous and Discrete Optimization”. He received his Ph.D. in electrical and systems engineering from the University of Pennsylvania (Penn). He is the recipient of the Army Research Office (ARO) Early Career Program Award, the Simons-Berkeley Research Fellowship, and Penn’s Joseph and Rosaline Wolf Award for Best Doctoral Dissertation.

Machine learning applications to structural dynamics and transportation policy

Machine learning techniques have the ability to automatically generate a model using data from past experiences. The number of applications is extensive, ranging from the automotive industry to house price estimations. In this talk, Dr. Chierichetti will present her research on applications of machine learning to transient stress predictions, as well as sensor placement for vibration testing. She will also present her experience with policy analysis in transportation, in which machine learning can be successfully used to understand citizens sentiment with respect to potential new laws, as well as to determine the effectiveness of existing policies in the transportation industry.

About Maria Chierichetti
Maria Chierichetti is an Associate Professor in Aerospace Engineering at San Jose' State University (CA). She received her MS and PhD in Aerospace Engineering from Georgia Institute of Technology, and BS and MS from Politecnico di Milano (Italy). She is an Amelia Earhart Fellow - Zonta International Foundation. Dr. Chierichetti's research focuses on the application of machine learning to structural dynamics, as well as on the analysis of the safety of transportation systems from a policy perspective, both on the ground and in the airspace. She is currently a research associate for the Mineta Transportation Institute. She has also been involved in several activities that promote students' success: she has researched how students learning and experience are affected by external factors, such as COVID-19 pandemic and community service. She has experience mentoring fellow engineering faculty on active learning, successful state-of-the-art teaching strategies pedagogies. Before joining SJSU, she worked as a faculty member at Worcester Polytechnic Institute and at the University of Cincinnati.

Long-run Performance of Approximation Algorithms

Professor Milgrom and his team study investment incentives created by truthful mechanisms that allocate resources using approximation algorithms. Even for some high-performing ("FPTAS") approximation algorithms, when a bidder can invest before participating, its investment incentives may be so distorted that the net welfare performance is arbitrarily bad. An algorithm’s worst-case allocation and investment performance coincide if and only if a particular kind of externality is sufficiently small. The researchers introduce a new FPTAS for the knapsack problem that has no such negative externalities, so it is high-performing with and without investments.

About the Paul Milgrom
Paul Milgrom is the Ely Professor of Humanities and Sciences in the Department of Economics at Stanford University. He is the recipient of numerous awards, including the 2020 Sveriges Riksbank Prize, in Memory of Alfred Nobel, for “improvements to auction theory and invention of new auction methods.”

Milgrom is the author of two books about auction design, and his scholarly publications have more than 100,000 Google Scholar citations. He co-invented the two auction formats most commonly used for selling radio spectrum licenses in North America, Europe, Asia, and Australia, and the Auctionomics team that designed the U.S. Incentive Auction process which reallocated UHF-TV channels for use in mobile broadband.

Assisted Learning: A Learning Framework for Organizations with Limited and Imbalanced Data

We develop an assisted learning framework for assisting organization-level learners in improving their learning performance with limited and imbalanced data. In particular, learners at the organizational level usually have sufficient computation resource, but are subject to stringent collaboration policy and information privacy. Their limited imbalanced data often cause biased inference and sub-optimal decision-making. In our assisted learning framework, an organizational learner purchases assistance service from a service provider and aims to enhance its model performance within a few assistance rounds. We develop effective stochastic training algorithms for assisted deep learning and assisted reinforcement learning. Different from existing distributed algorithms that need to frequently transmit gradients or models, our framework allows the learner to only occasionally share information with the service provider, and still achieve a near-oracle model as if all the data were centralized.

About Yi Zhou
Yi Zhou is an assistant professor affiliated with the Department of ECE at the University of Utah. Before, he worked as a post-doctorate research associate in the Department of ECE at Duke University. He obtained a Ph.D. degree in ECE from The Ohio State University in 2018. His research interests include deep learning, reinforcement learning, statistical machine learning, nonconvex and distributed optimization, and statistical signal processing.

Peptide targeting for novel therapies and cell biology

Dr. Pang is an assistant professor at the Department of Pharmaceutics, University of Minnesota. He graduated from University of Utah as a PhD of biochemistry, and did postdoc training at Sanford-Burnham-Prebys institute at La Jolla, CA.

His lab is specialized in developing peptide-based targeted therapeutics, and eliciting the cellular transport processes. Using phage display, his lab is screening for peptides targeting any given disease or cell type, and deciphering the underlying biomarkers. Through chemical modification and nanotechnology, his lab applies targeting peptides to improve the delivery efficiency and specificity of therapeutic and diagnostic payloads. Moreover, he is interested in the endocytosis, exocytosis and intercellular communication triggered by peptides and their payloads. Using simulation and experimental approaches, his group is investigating the biophysical and molecular basis of these transport processes.

Join industry thought leaders for a panel discussion on data science hosted by CEMS on data science and its impact in industry..

Hear perspectives on:

• Industry workforce needs
• Current and future deployment of data science tools and practices
• Success stories and barriers
• Major trends and open problems
• Industry-academia engagement opportunities

• Current and future deployment of data science tools and practices, along with a new MS degree program bridging chemical engineering and materials science with data science

Panelists include:

• Dr. Leo Chiang, R&D Fellow in AI and Data Science, Dow
• Dr. Michael Dolezal, Vice President of R&D, 3M Digital Science Community
• Dr. Bryce Meredig, Chief Science Officer and Co-Founder, Citrine Informatics
• Dr. Fernando Ulloa Montoya, Head, Data and Computational Science, mRNA Center of Excellence, Sanofi
• Dr. Upendra Ummethala, Vice President, Applied AIx Products, Applied Materials