Eventshttp://www.cam.cornell.eduEventsWed, 19 Sep 2018 18:28:33 -0400CAM Colloquium: Anil Damle (CS, Cornell University) - A unified approach to Wannier localization: building localized basis functions and interpolating eigenvaluesAbstract: The so-called Wannier localization problem in quantum physics is analogous to finding a localized representation of a subspace associated with a nonlinear eigenvalue problem and plays an important role in Hartree-Fock and Kohn-Sham density functional theory calculations. While this problem is well studied for insulating systems and good algorithms exist, considerably less is known for metallic systems with entangled eigenvalues. We propose a new, unified method to solve the Wannier localization problem that works in both the isolated and entangled setting. Our method is robust, direct, efficient, and does not require an initial guess. We will demonstrate the effectiveness of our methodology at constructing localized basis functions that may subsequently be used for Wannier interpolation of band structure. Bio: Anil Damle is an assistant professor of Computer Science at Cornell University where he works at the intersection of numerical linear algebra and a broad range of application areas. His work includes the development of efficient and robust algorithms that leverage the underlying physical and/or statistical structure of problems. Anil received his BS and MS from the University of Colorado, Boulder in 2011 and his Ph.D. from Stanford University in 2016.http://www.cam.cornell.edu/news/colloquium.cfm?event=18923
http://www.cam.cornell.edu/news/colloquium.cfm?event=18923
Fri, 21 Sep 2018 15:30:00 -0400CAM Colloquium: Jayadev Acharya (ECE, Cornell University) - Statistical inference under local information constraintsAbstract: Independent samples from an unknown probability distribution p on a domain of size k are distributed across n players, with each player holding one sample. Each player can communicate L bits to a central referee, with the goal of resolving a prespecified inference problem. When L >= log k bits, the problem reduces to the well-studied centralized case, where all the samples are available in one place. We focus on the communication-starved setting L < log k, in which as we will see the landscape changes drastically. We develop a general formulation for inference problems in this distributed setting, and instantiate it for two prototypical inference questions, learning and identity testing. We will consider and discuss the power of shared randomness in distributed inference, and show that for identity testing, schemes without public randomness can be dramatically less efficient than those with. Finally, we show that our framework is general enough to be extended to other distributed settings, in particular to local differential privacy. Based on joint works with Clement Canonne, Cody Freitag, and Himanshu Tyagi. Bio: Acharya joined Cornell in August 2016 as an Assistant Professor in the school of Electrical and Computer Engineering, after spending two years as a postdoc at MIT. He obtained his Ph.D. from University of California, San Diego, and his B. Tech degree from Indian Institute of Technology, Kharagpur.http://www.cam.cornell.edu/news/colloquium.cfm?event=18926
http://www.cam.cornell.edu/news/colloquium.cfm?event=18926
Fri, 28 Sep 2018 15:30:00 -0400CAM Colloquium: No Colloquium - Fall BreakFall Breakhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18844
http://www.cam.cornell.edu/news/colloquium.cfm?event=18844
Fri, 05 Oct 2018 15:30:00 -0400CAM Colloquium: Douglas Down (McMaster University) - Energy-Aware Scheduling on Heterogeneous ServersAbstract: This talk is concerned with the scheduling of a multiserver system where servers are heterogeneous in terms of both their speeds and their usage (energy) costs. The goal is to determine scheduling policies that minimize a combination of holding costs (cost for waiting) of tasks and energy/usage costs of servers. The motivation is designing task scheduling policies for reducing energy consumption in data centers. We discuss some previous results on determining optimal scheduling policies for clearing systems (there are no arrivals) and systems with arrivals as well as whether tasks can be reassigned or not. A key step in determining optimality of a scheduling policy is proving relations between an individually optimal (greedy) scheduling policy and a socially optimal scheduling policy. This relation will be discussed in some detail. Finally, two recent extensions will be presented - incorporating a thermal model for servers and service rate control. Bio: Douglas Down is a Professor in the Department of Computing and Software and the Acting Academic Director of the Computing Infrastructure Research Centre, both at McMaster University in Hamilton, Ontario, Canada. He received his PhD in Electrical Engineering from the University of Illinois at Urbana-Champaign. His research interests lie in the area of queueing theory, performance modelling of computer systems and control systems design for data centers.http://www.cam.cornell.edu/news/colloquium.cfm?event=18883
http://www.cam.cornell.edu/news/colloquium.cfm?event=18883
Fri, 12 Oct 2018 15:30:00 -0400CAM Colloquium: Joel Tropp (California Institute of Technology) - Applied random matrix theoryAbstract: Random matrices now play a role in many areas of theoretical, applied, and computational mathematics. Therefore, it is desirable to have tools for studying random matrices that are flexible, easy to use, and powerful. Over the last fifteen years, researchers have developed a remarkable family of results, called matrix concentration inequalities, that balance these criteria. This talk offers an invitation to the field of matrix concentration inequalities and their applications. Bio: Joel A. Tropp is the Steele Family Professor of Applied & Computational Mathematics at the California Institute of Technology. He earned a Ph.D. degree in Computational Applied Mathematics from the University of Texas at Austin in 2004. His research centers on data science, applied mathematics, numerical algorithms, and random matrix theory. Prof. Tropp won a PECASE in 2008, and he has received society best paper awards from SIAM in 2010, EUSIPCO in 2011, and IMA in 2015. He has also been recognized as a Highly Cited Researcher in Computer Science each year from 2014–2017.http://www.cam.cornell.edu/news/colloquium.cfm?event=18908
http://www.cam.cornell.edu/news/colloquium.cfm?event=18908
Fri, 19 Oct 2018 15:30:00 -0400CAM Colloquium: Laurent Lessard (University of Wisconsin-Madison) - TBDLaurent Lessard - TBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18878
http://www.cam.cornell.edu/news/colloquium.cfm?event=18878
Fri, 26 Oct 2018 15:30:00 -0400CAM Colloquium: Suriya Gunasekar (Toyota Technological Institute, Chicago) - TBDSuriya Gunasekar - TBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18899
http://www.cam.cornell.edu/news/colloquium.cfm?event=18899
Fri, 02 Nov 2018 15:30:00 -0400CAM Colloquium: Cecilia Diniz Behn (Colorado School of Mines & University of Colorado School of Medicine) - Mapping napping: a map-based approach to understanding sleep dynamics in early childhoodAbstract: The temporal structure of human sleep changes across development as sleep consolidates from the fragmented sleep of infants to the single nighttime sleep period typical in adults. This transition is likely driven by developmentally-mediated slowing in the rates of accumulation and dissipation of the homeostatic sleep drive, the physiological drive to sleep that increases with time spent awake. Experimental advances in identifying the neurophysiology of sleep-wake regulation have led to the formulation of physiologically-based mathematical models that describe the dynamics of neuronal interactions within the sleep-wake regulatory network and their modulation by homeostatic and circadian drives. Given the periodicity of regular, entrained sleep-wake behavior, one-dimensional circle maps—defined analytically in some cases and numerically in others—may be used to represent the dynamics of the full sleep-wake network model. In this talk I will describe mathematical modeling of sleep, the piecewise continuous maps that represent model dynamics, and bifurcations arising in these maps. Specifically, tracking bifurcations as homeostatic sleep drive time constants are varied provides insight into the structure underlying the consolidation of sleep from many to one sleep period per day. This analysis has implications for understanding the dynamics of the transition from napping to non-napping behavior in early childhood. Bio: Cecilia Diniz Behn is an Assistant Professor in the Department of Applied Mathematics and Statistics at Colorado School of Mines. She also holds an appointment as an Adjoint Assistant Professor of Pediatrics at the University of Colorado School of Medicine. She received her PhD in Mathematics from Boston University and conducted postdoctoral work at Harvard Medical School and the University of Michigan. Her research focuses on dynamical systems and multiscale mathematical modeling with applications in metabolism, sleep, and circadian rhythms.http://www.cam.cornell.edu/news/colloquium.cfm?event=18921
http://www.cam.cornell.edu/news/colloquium.cfm?event=18921
Fri, 09 Nov 2018 15:30:00 -0400CAM Colloquium: Drew Kouri (Sandia National Laboratories) - A primal-dual algorithm for large-scale risk minimizationAbstract: Many science and engineering applications necessitate the optimization of systems described by partial differential equations (PDEs) with uncertain inputs including noisy physical parameters, unknown boundary or initial conditions, and unverifiable modeling assumptions. One can formulate such problems as risk-averse optimization problems in Banach space, which upon discretization, become enormous risk-averse stochastic programs. For many popular risk models including the coherent risk measures, the resulting risk-averse objective function is not differentiable. This lack of differentiability complicates the numerical approximation of the objective function as well as the numerical solution of the optimization problem. To address these challenges, I present a general primal-dual algorithm for solving large-scale nonsmooth risk-averse optimization problems. This algorithm is motivated by epigraphical regularization of risk measures and is closely related to the classical method of multipliers. As a result, the algorithm solves a sequence of smooth optimization problems using derivative-based methods. I prove convergence of the algorithm even when the subproblem solves are performed inexactly and conclude my presentation with multiple PDE-constrained examples that demonstrate the efficiency of this method. Bio: Drew Kouri is a staff member in the Optimization and Uncertainty Quantification Department at Sandia National Laboratories. He received his BS and MS (2008) in mathematics from Case Western Research University, and his MA (2010) and PhD (2012) in computational and applied mathematics from Rice University. Before joining Sandia, he was the J. H. Wilkinson Fellow at Argonne National Laboratory. His research focuses on the analysis and numerical solution of PDE-constrained optimization and stochastic programming problems.http://www.cam.cornell.edu/news/colloquium.cfm?event=18941
http://www.cam.cornell.edu/news/colloquium.cfm?event=18941
Fri, 16 Nov 2018 15:30:00 -0400CAM Colloquium: No Colloquium - Thanksgiving BreakThanksgiving Breakhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18871
http://www.cam.cornell.edu/news/colloquium.cfm?event=18871
Fri, 23 Nov 2018 15:30:00 -0400CAM Colloquium: TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18850
http://www.cam.cornell.edu/news/colloquium.cfm?event=18850
Fri, 30 Nov 2018 15:30:00 -0400CAM Colloquium: Andy Borum (Mathematics, Cornell) - TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18965
http://www.cam.cornell.edu/news/colloquium.cfm?event=18965
Fri, 25 Jan 2019 15:30:00 -0400CAM Colloquium: TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18852
http://www.cam.cornell.edu/news/colloquium.cfm?event=18852
Fri, 01 Feb 2019 15:30:00 -0400CAM Colloquium: TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18853
http://www.cam.cornell.edu/news/colloquium.cfm?event=18853
Fri, 08 Feb 2019 15:30:00 -0400CAM Colloquium: Jiahao Chen (Capital One, New York) - TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18966
http://www.cam.cornell.edu/news/colloquium.cfm?event=18966
Fri, 15 Feb 2019 15:30:00 -0400CAM Colloquium: No Colloquium - February BreakFebruary Breakhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18870
http://www.cam.cornell.edu/news/colloquium.cfm?event=18870
Fri, 22 Feb 2019 15:30:00 -0400CAM Colloquium: TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18858
http://www.cam.cornell.edu/news/colloquium.cfm?event=18858
Fri, 01 Mar 2019 15:30:00 -0400CAM Colloquium: Jie Shen (Purdue) - TBDJie Shen - TBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18877
http://www.cam.cornell.edu/news/colloquium.cfm?event=18877
Fri, 08 Mar 2019 15:30:00 -0400CAM Colloquium: TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18860
http://www.cam.cornell.edu/news/colloquium.cfm?event=18860
Fri, 15 Mar 2019 15:30:00 -0400CAM Colloquium: TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18861
http://www.cam.cornell.edu/news/colloquium.cfm?event=18861
Fri, 22 Mar 2019 15:30:00 -0400CAM Colloquium: No Colloquium - Spring BreakSpring Breakhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18835
http://www.cam.cornell.edu/news/colloquium.cfm?event=18835
Fri, 29 Mar 2019 15:30:00 -0400CAM Colloquium: No Colloquium - Spring BreakSpring Breakhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18836
http://www.cam.cornell.edu/news/colloquium.cfm?event=18836
Fri, 05 Apr 2019 15:30:00 -0400CAM Colloquium: Emery Brown (Harvard/MIT) - TBDEmery Brown - TBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18876
http://www.cam.cornell.edu/news/colloquium.cfm?event=18876
Fri, 12 Apr 2019 15:30:00 -0400CAM Colloquium: Lin Lin (UC Berkeley) - TBDLin Lin - TBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18911
http://www.cam.cornell.edu/news/colloquium.cfm?event=18911
Fri, 19 Apr 2019 15:30:00 -0400CAM Colloquium: Todd Coleman (UCSD) - TBDTodd Coleman - TBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18869
http://www.cam.cornell.edu/news/colloquium.cfm?event=18869
Fri, 26 Apr 2019 15:30:00 -0400CAM Colloquium: Afonso Bandeira (Courant, NYU) - TBDTBDhttp://www.cam.cornell.edu/news/colloquium.cfm?event=18868
http://www.cam.cornell.edu/news/colloquium.cfm?event=18868
Fri, 03 May 2019 15:30:00 -0400