Dm-meeting

Download

dm-meeting@cs.vt.edu

March 2019

2 participants
2 discussions

Fwd: [Dac-faculty] Fwd: ECE- FACULTY FACULTY Faculty candidate seminar - Machine Perception for Autonomous Systems
by B. Aditya Prakash 26 Mar '19

26 Mar '19

Seems interesting! You guys should attend. Btw, also this week there are talks by Jen Neville and Mike Jordan; you should definitely attend those. There is a student meeting with Jen too; I would suggest you try to make it for that too, if possible. Cheers, - Aditya Apologies for the brevity Sent from my iPhone Begin forwarded message: > From: Jia-Bin Huang via Dac-faculty <dac-faculty(a)cs.vt.edu> > Date: March 26, 2019 at 7:01:27 PM EDT > To: "dac-faculty(a)cs.vt.edu" <dac-faculty(a)cs.vt.edu> > Subject: [Dac-faculty] Fwd: ECE- FACULTY FACULTY Faculty candidate seminar - Machine Perception for Autonomous Systems > Reply-To: Jia-Bin Huang <jbhuang(a)vt.edu> > > Dear Colleagus, > > I will be hosting a faculty seminar for the ECE machine perception position. As the candiate’s research overlaps with several of the current DAC faculty members (machine learning, social networks, data mining, graph signal processing), I am forwarding the talk information and hope to see you and your students could attend the talk. > > Time: 10:00 AM - 11:00 AM March 28 > Location: 457B Whittemore > > Thank you very much! > > Best, > > Jia-Bin > > ---------- Forwarded message --------- > From: Jia-Bin Huang <jbhuang(a)vt.edu> > Date: Tue, Mar 26, 2019 at 12:00 PM > Subject: ECE- FACULTY FACULTY Faculty candidate seminar - Machine Perception for Autonomous Systems > To: <ece-faculty-g(a)vt.edu> > > > Dear Colleagues, > > You are invited to a presentation this Thursday on the following topic by Dr. Siheng Chen (Research Scientist at MERL). > > Time: 10:00 AM - 11:00 AM March 28 > Location: 457B Whittemore > > Title: Data science with graphs: From social network analysis to autonomous driving > > Abstract: A massive amount of data is being generated at an unprecedented level from a diversity of sources, including online social networks, cyber-physical systems, biological networks, urban traffic networks, 3D point clouds, and many others. These seemingly different types of data share a similar pattern: data are associated with complex and irregular structures. The necessity of analyzing such data has led to the birth of a unifying framework, data science with graphs. This framework offers a mathematically rigorous paradigm to analyze high-dimensional data associated with complex and irregular graph structures. It extends tools and concepts from classical signal processing and machine learning to the graph domain. In this talk, we will introduce this emerging framework and consider a series of graph-based theories and methodologies, including sampling theory of graph data and graph-based deep neural networks. We will illustrate the power of the proposed methodologies with the applications to social network analysis, urban spatial-temporal data analysis, and autonomous driving. > > Bio: Siheng Chen is a research scientist at Mitsubishi Electric Research Laboratories. Before that, he was an autonomy engineer at Uber Advanced Technologies Group, working on the perception and prediction systems of self-driving cars. Before joining Uber, Dr. Chen was a postdoctoral research associate at Carnegie Mellon University. Dr. Chen received his doctorate in Electrical and Computer Engineering from Carnegie Mellon University in 2016, where he also received two masters degrees in Electrical and Computer Engineering (College of Engineering) and Machine Learning (School of Computer Science), respectively. He received his bachelor’s degree in Electronics Engineering in 2011 from Beijing Institute of Technology, China. He is the recipient of the 2018 IEEE Signal Processing Society Young Author Best Paper Award. His coauthored paper received the Best Student Paper Award at 2018 IEEE Global Conference on Signal and Information Processing. His research interests include signal processing, machine learning, 3D point cloud processing, and autonomous driving. > > Best, > > Jia-Bin Huang > https://filebox.ece.vt.edu/~jbhuang/ > > -- > You received this message because you are subscribed to the Google Groups "ECE faculty (ece-faculty-g)" group. > To unsubscribe from this group and stop receiving emails from it, send an email to ece-faculty-g+unsubscribe(a)vt.edu. > To post to this group, send email to ece-faculty-g(a)vt.edu. > _______________________________________________ > Dac-faculty mailing list > Dac-faculty(a)cs.vt.edu

1 0

Wednesday, March 20th @ 4 PM, McB 133C, Xinfeng Xu
by Alexander Rodriguez 18 Mar '19

18 Mar '19

Please join us for our next DM-Meeting. ****DM-Meeting Spring 2019**** WHO: Xinfeng Xu WHEN: Wednesday, March 20th @ 4 PM WHERE: McB 133C WHAT: Xinfeng will practice his master thesis defense "" Title: Modeling and Predicting Incidence: Critical Systems Failures and Flu Infection Cases Abstract: Given several related critical infrastructure (CI) networks, such as power grid, transportation, and water systems, one crucial question emerges: how to model the propagation of failed facilities and predict their spread over time to the whole system? Given digital surveillance data, can we predict the impact of Influenza-Like Illness (ILI), including the percentage of outpatient doctor visits, the season duration, and peak? These questions are related to modeling and predicting the incidence of different types of contagions. In the case of CI, the contagions are the failures of facilities. In the case of flu spread, the contagions are the infective ILIs. In this thesis, in the case of CI, we give a novel model of failure cascades and use it to identify key facilities in an optimization-based approach, called HotSpots. In the case of flu spread, we develop a deep neural network, EpiDeep, to predict multiple key epidemiology metrics. In both of these applications, we use the dynamics of propagation to develop better algorithms.\\ By collaborating with Oak Ridge National Laboratory (ORNL) and working on the real CI networks provided by them, we find that HotSpots helps solve what-if scenarios. By carrying on experiments on the digital surveillance data from the Centers for Disease Control and Prevention (CDC), we find that EpiDeep is better than non-trivial baselines and outperforms them by up to 40\%. We believe that our results are general and can be applied to other propagation-based scenarios in infrastructure and epidemiology. Best regards, Alex _______________________________________________ Seminar website: *http://people.cs.vt.edu/~arodriguez/DM-Meeting/home.html <http://people.cs.vt.edu/~arodriguez/DM-Meeting/home.html>* Please contact Alexander Rodriguez (arodriguez(a)cs.vt.edu <esorour(a)cs.vt.edu>), if interested in giving a talk. Dm-meeting mailing list Dm-meeting(a)cs.vt.edu https://mailman.cs.vt.edu/mailman/listinfo/dm-meeting

1 0