Events for the 1st week of November
-
Center for Systems and Control (CSC@USC) and Ming Hsieh Institute for Electrical Engineering
Mon, Oct 30, 2017 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Anil Aswani, University of California, Berkeley
Talk Title: Making Robust Decisions from Data
Series: Fall 2017 Joint CSC@USC/CommNetS-MHI Seminar Series
Abstract: Though machine learning has found success in decision-making contexts, these methods are fragile to model mismatch and malicious interference. This is a major impediment to the deployment of automated decision-making in safety-critical systems like those found in healthcare or physical infrastructure. This talk describes three methods we have developed for robust decision-making in different scenarios. The first is a framework for combining robust control with machine learning, and applications to energy efficiency and robotics are highlighted. The second is algorithms to solve inverse optimization (and inverse reinforcement learning) with noisy data. This problem arises when estimating utility functions or modeling human-automation systems, and we show it is NP-hard and that existing approaches are statistically inconsistent. We develop a polynomial time algorithm that is asymptotically optimal as more data is collected. Then we discuss applications of our inverse optimization approach to a clinical trial on personalized goal-setting through smartphone apps to increase physical activity, and to studying an incentive design problem in the Medicare Shared Savings Program where we show that an investment sharing plan could potentially save Medicare an additional $85 million per year. The third is an approach for bandit models where repeated application of an action causes habituation and a decrease of that action's rewards, while refraining from an action causes recovery and an increase of that action's awards. Though such problems are PSPACE-complete, we define a class of models called ROGUE bandits for which we can construct policies that achieve logarithmic regret. We describe an application of ROGUE bandits to a personalized healthcare problem of choosing an optimal sequence of daily messages to encourage an individual to increase their physical activity.
Biography: Anil Aswani is an Assistant Professor in Industrial Engineering and Operations Research (IEOR) at UC Berkeley. He received his B.S. in Electrical Engineering from the University of Michigan in 2005, M.S. in Electrical Engineering and Computer Sciences (EECS) from UC Berkeley in 2007, and Ph.D. in EECS from UC Berkeley in 2010. He received a Hellman Fellowship for his research on food insecurity, the Leon O. Chua award from Berkeley for outstanding achievement in an area of nonlinear science, and a William Pierskalla Runner-Up Award from the INFORMS Health Applications Society. His research interests include data-driven decision making, with particular emphasis on addressing inefficiencies and inequities in health systems and physical infrastructure.
Host: Insoon Yang, insoonya@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
Audiences: Everyone Is Invited
Contact: Gerrielyn Ramos
-
Ming Hsieh Institute Seminar Series on Integrated Systems
Fri, Nov 03, 2017 @ 10:30 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Jane Gu, Associate Professor, University of California, Davis
Talk Title: Phase Noise Filter for LO Phase Noise Suppression
Host: Profs. Hossein Hashemi, Mike Chen, Mahta Moghaddam, and Dina El-Damak
More Information: MHI Seminar Series IS - Jane Gu.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Jenny Lin
-
Munushian Speaker - Ana Claudia Arias, Friday, November 3rd at 2:00pm in EEB 132
Fri, Nov 03, 2017 @ 02:00 PM - 03:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Ana Claudia Arias, University of California, Berkeley
Talk Title: Printed Flexible Wearable Medical Devices
Abstract: Wearable noninvasive medical sensing is extremely promising for monitoring human performance during physically demanding tasks. Printed sensors provide a distinct advantage over rigid sensors at establishing high-fidelity sensor-skin interfaces due to their inherently flexible material systems and form factors. Hence, these sensors are suitable for monitoring vital signs as well as analytes in bodily fluids. We have demonstrated an integrated wearable and flexible multi-sensor platform capable of simultaneous bioelectronic and biophotonic sensing of physiological state of the human body. The sensor platform is
composed of printed photoplethysmography (PPG) and sweat sensors. The PPG sensor uses blade coated red and near-infrared (NIR) organic light-emitting diodes (OLEDs) and an organic photodiode (OPD) to optically measure heart rate and blood oxygenation. The sweat sensor is comprised of printed sodium, ammonium and lactate sensors. The lactate sensor exhibits high sensitivity of 26 uA/mM cm2 with the linear detection range of 1-20 mM of lactate. The sodium and ammonium sensors exhibit near-Nernstian response with sensitivities of 60.0 +/- 4.0 mV (detection range: 1-100 mM) and 56.1 +/- 2.2mV (detection range: 0.1-100 mM) per decade of concentrations, respectively. The sensors are interfaced with a Bluetooth System on Chip for wirelessly reading out sensor data. The complete system is powered by two flexible printed batteries at 8 V and are of 40 mAh capacity. This integrated platform can provide meaningful data to the end-users or healthcare professionals stretching the application domain of wearable sensing beyond the fitness domain to medical diagnostics.
Biography: Dr. Ana Claudia Arias is a Professor at the Electrical Engineering and Computer Sciences Department at the University of California in Berkeley and a faculty director at the Berkeley Wireless Research Center (BWRC) and SWARM lab. Prior to joining the University of California she was the head of the Printed Electronic Devices Area and a Member of Research Staff at PARC, a Xerox Company, Palo Alto, CA. She went to PARC from Plastic Logic in Cambridge, UK where she led the semiconductor group. She received her PhD on semiconducting polymer blends for photovoltaic devices from the Physics Department at the University of Cambridge, UK. Prior to that, she received her master and bachelor degrees in Physics from the Federal University of Paraná in Curitiba, Brazil. Her research focuses on devices based on solution processed materials and applications development for flexible sensors and electronic systems. Ana Claudia is the chair of ThinFilm Electronics Technical Advisory Council and she is an author of over 100 peer reviewed publications and issued patents.
Host: EE-Electrophysics
More Info: minghsiehee.usc.edu/about/lectures/munushian
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
Event Link: minghsiehee.usc.edu/about/lectures/munushian
-
Electrical Engineering Seminar
Fri, Nov 03, 2017 @ 03:00 PM - 04:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr Vincent Gripon, IMT Atlantique
Talk Title: Generalizing Convolutional Neural Networks to Graph Domains
Abstract: For the past few years, various works have aimed at tackling the problem of extending Convolutional Neural Networks (CNNs)to irregular domains. In this presentation, we propose to use graphs as an intermediate solution, thus deriving two subproblems: a) identifying a graph given a set of signals and b) defining CNN-like structures given
a graph domain. For both problems we introduce original approaches and discuss their performance.
Biography: Vincent Gripon is a permanent researcher with IMT-Atlantique (Institut Mines-Télécom), Brest, France. He obtained his M.S. from Ãcole Normale Supérieure of Cachan and his Ph.D. from Télécom Bretagne. His research interests lie at the intersection of information
theory, computer science and neural networks. He co-authored about 60 papers in the above-mentioned domains.
Host: Dr. Antonio Ortega
More Information: Vincent Gripon Seminar.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Gloria Halfacre