Select a calendar:
Filter January Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Events for January 27, 2015
-
CS Colloquium: Dr. Chris Paredis (NSF) - A Theoretical Framework for Systems Engineering and Design - Asking "Why?" rather than just "How?"
Tue, Jan 27, 2015 @ 11:00 AM - 12:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dr. Chris Paredis, National Science Foundation
Talk Title: A Theoretical Framework for Systems Engineering and Design - Asking "Why?" rather than just "How"?
Series: CS Colloquium
Abstract: An important component of the Engineering and Systems Design (ESD) and Systems Science (SYS) programs at the National Science Foundation is the development of a theoretical foundation for systems engineering and design. In this presentation, I will motivate the need for a theoretical framework and explain what it is. As an example, a model for systems engineering and design will be introduced in which the artifact development process is modeled as a search process. Unlike many models in the literature, this model provides an explanation for (rather than just a description of) current practices. Related to the need for a theoretical framework is the need for rigorous research methodology. I will therefore present some thoughts on the desired characteristics of a rigorous research methodology for systems engineering and design. The presentation will end with some logistical details about the NSF programs and an opportunity for Q&A. The overall goal is to increase the success of principal investigators when submitting proposals to the ESD or SYS programs.
Biography: Dr. Chris Paredis is Program Director for the Engineering and Systems Design (ESD) and Systems Science (SYS) programs at the National Science Foundation. He is also Professor of Mechanical Engineering in the G.W. Woodruff School of Mechanical Engineering, and in the H.M. Stewart School of Industrial and Systems Engineering at Georgia Tech, Atlanta, USA. He is a Woodruff Faculty Fellow and Director of the Model-Based System Engineering Center. He holds graduate degrees in Mechanical Engineering from the Catholic University of Leuven (Belgium) and in Electrical and Computer Engineering from Carnegie Mellon University.
Dr. Paredis' research focuses on Model-Based Systems Engineering, combining aspects of decision theory, information technology, simulation, and systems theory to support the design of complex mechatronic systems. In these areas, he has published more than 140 refereed book chapters, journal articles and conference papers. He has made significant contributions to the development of the theory, methods and tools for decision making in design and systems engineering, and has developed several decision support tools using the SysML and Modelica languages. As Director of the Model-Based Systems Engineering Center, he leads a group of 11 faculty members towards the development of a next generation of systems engineering methods that are value-driven, model-based and human-centered.
Within the Object Management Group (OMG), he served on the SysML Revision Task Force and has led the development of the SysML-Modelica Transformation Specification. He was a founding board member of the North America Modelica Users' Group, and served on the working group responsible for the INCOSE "Systems Engineering Vision 2025." He is also a member of the Technical Advisory Board for Integrated Model-Centric Engineering Program at NASA/JPL, is past Chair of the ASME Computers and Information in Engineering division, and has served as Conference Chair for the 2013 Conference on Systems Engineering Research (CSER'13) and the 2007 Computers and Information in Engineering Conference (CIE'07).
Dr. Paredis has served as Associate Editor for the SAE Journal of Commercial Vehicles and the ASME Journal of Mechanical Design. He currently serves as co-Editor of the ASME book series, "Advances in Computers and Information in Engineering Research." He received the 2007 CETL/BP Junior Faculty Teaching Excellence Award, the 2007 SAE Ralph R. Teetor Educational Award, and the 2011 ASME CIE Excellence in Research Award.
Host: CSSE
Location: Auditorium
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium Lecture Series: Rich Caruana (Microsoft Research) - Do Deep Nets Really Need To Be Deep?
Tue, Jan 27, 2015 @ 04:00 PM - 05:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Rich Caruana , Microsoft Research
Talk Title: Do Deep Nets Really Need To Be Deep?
Series: CS Colloquium
Abstract: Currently, deep neural networks are the state of the art on problems such as speech recognition and computer vision. By using a method called model compression, we show that shallow feed-forward nets can learn the complex functions previously learned by deep nets and achieve accuracies previously only achievable with deep models while using the same number of parameters as the original deep models. On the TIMIT phoneme recognition and CIFAR-10 image recognition tasks, shallow nets can be trained that perform similarly to complex, well-engineered, deeper convolutional architectures. The same model compression trick can also be used to compress impractically large deep models and ensembles of large deep models down to âmedium-sizeâ deep models that run more efficiently on servers, and down to âsmallâ models that can run on mobile devices. In machine learning and statistics we used to believe that one of the keys to preventing overfitting was to keep models simple and the number of parameters small to force generalization. We no longer believe this --- learning appears to generalize best when training models with excess capacity, but the learned functions can often be represented with far fewer parameters.
The lecture can be streamed HERE
Host: Yan Liu
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
