Date/Time : 12 Feb 2019 14:00
Venue : S17 #04-06 SR1
Speaker : Shigeru Kanemitsu, Kyushu Institute of Technology

Add to calendar :

Date/Time : 11 Feb 2019 15:00
Venue : S17 #04-06 SR1
Speaker : Yerkin Kitapbayev, Sloan School of Management, MIT

Date/Time:11 Feb 2019 15:00 Venue: S17 #04-06 SR1 Speaker: Yerkin Kitapbayev, Sloan School of Management, MIT Optimal investment strategies for power generation: the value of green energy This paper examines the investment in and the valuation of power generation projects under uncertainty. The analysis incorporates the possibility of producing from alternative types of fuels, such as renewables (wind) or fossil fuels (gas), hence alternative types of plants/technologies. The model considered in this paper cannot be reduced to a single state variable. It involves two distinct underlying processes, for electricity and gas, as well as embedded (two-dimensional) optimal stopping time problems. We apply probabilistic arguments and show that the optimal investment decision can be characterized by two boundaries that satisfy a system of coupled integral equations of Fredholm type. The value of a project to invest in the best of the two types of plants is the premium associated with the optimal investment policy, i.e., the present value of the cumulative instantaneous gains realized when either of the two technologies is optimal. The ability to select the type of plant/technology has significant effects on investment in power generation. The optimal delay can be substantial even if the values of both projects are both large. Green power generation emerges as a significant source of value creation. (Joint work with Jerome Detemple). Add to calendar: 

Add to calendar :

Date/Time : 31 Jan 2019 15:00
Venue : S17 #05-11 SR5
Speaker : Khoo Yuehaw, Stanford University

Date/Time:31 Jan 2019 15:00 Venue: S17 #05-11 SR5 Speaker: Khoo Yuehaw, Stanford University Multimarginal optimal transport, density functional theory, and convex relaxation Density functional theory is an effective tool in solid state physics and quantum chemistry for electronic structure calculation. However, it has difficulties when dealing with strongly correlated systems. In this talk, we examine the regime where the electrons are strictly correlated. This gives rise to a multimarginal optimal transport problem, a direct extension of the optimal transport problem that has applications in economics and operations research as well. In particular we introduce methods from convex optimization to provide a lower bound to the cost of the multimarginal transport problem with a practical running time. We further propose projection schemes based on tensor decomposition to obtain upper bounds to the energy. Numerical experiments demonstrate a gap of order $10^{-3}$ to $10^{-2}$ between the upper and lower bounds. Add to calendar: 

Add to calendar :

Date/Time : 28 Jan 2019 15:00
Venue : S17 #04-06 SR1
Speaker : Zhang Zhiyuan, Institute for Advanced Study

Date/Time:28 Jan 2019 15:00 Venue: S17 #04-06 SR1 Speaker: Zhang Zhiyuan, Institute for Advanced Study Zimmer’s conjecture for simple complex Lie groups We will discuss Zimmer’s conjecture on the actions of lattices in higher rank Lie groups on manifolds of small dimension. This conjecture is recently proved by Aaron Brown, David Fisher and Sebastian Hurtado for real split simple Lie groups. We discuss the proof of this conjecture for simple complex Lie groups. We also give significant improvements for the dimensional bounds for many other non-split Lie groups. This is a joint work with Jinpeng An and Aaron Brown. Add to calendar: 

Add to calendar :

Date/Time : 17 Jan 2019 15:00
Venue : S17 #04-05 SR2
Speaker : Lee Ching-pei, University of Wisconsin-Madison

Date/Time:17 Jan 2019 15:00 Venue: S17 #04-05 SR2 Speaker: Lee Ching-pei, University of Wisconsin-Madison Convergence Rates for some Fundamental Algorithms for Continuous Minimization of functions that are a sum of a smooth part and a convex but possibly nonsmooth part that promotes regularization is prevalent in many ap- plication areas, and proximal-type methods are the most eective algorithms for such regularized optimization problems. In this talk, I will present our re- cent advancements on the convergence analysis of rst-order and second-order proximal-type methods for regularized optimization in [1, 2]. I will rst show that the long-known O(1/k) convergence rate on convex problems of gradient descent and coordinate descent in the function value is not tight and can be improved to o(1/k), and extend this result to proximal gradient and proximal coordinate descent for regularized optimization. The second part of this talk will focus on a general framework of inexact successive quadratic approxima- tion that has second-order proximal methods including proximal Newton and proximal quasi-Newton as special cases. Global convergence rates for strongly convex, convex, and nonconvex problems will be discussed. References: [1] Ching-pei Lee and Stephen J. Wright. First-order algorithms converge faster than O(1=k) on convex problems. Technical report, December 2018. URL http://www.optimization-online.org/DB_HTML/2018/12/6993.html. [2] Ching-pei Lee and Stephen J. Wright. Inexact successive quadratic approx- imation for regularized optimization. Computational Optimization and Ap- plications, 2019+. URL http://www.optimization-online.org/DB_HTML/ 2018/03/6499.html. Accepted. Add to calendar: 

Add to calendar :

Date/Time : 15 Jan 2019 15:00
Venue : S17 #04-05 SR2
Speaker : Leonard Huang , University of Nevada, Reno, USA

Add to calendar :

Date/Time : 30 Nov 2018 15:00
Venue : S17 #04-06 SR1
Speaker : Yuval Peres, Microsoft Research

Date/Time:30 Nov 2018 15:00 Venue: S17 #04-06 SR1 Speaker: Yuval Peres, Microsoft Research Search games and Kakeya sets I’ll describe a search game with a surprising geometric connection. A hunter and a rabbit move on an n-vertex graph without seeing each other until they meet. At each step, the hunter moves to a neighboring vertex or stays in place, while the rabbit is free to jump to any node. Thus they are engaged in a zero sum game, where the payoff is the capture time. We show that an optimal rabbit strategy for the cycle yields a Kakeya set: a plane set of zero area that contains a unit segment in every direction. Kakeya sets have been studied intensively in harmonic analysis since their discovery by Besicovitch (1919); their connection to search games is new and yields insights in both directions. (Based on joint work with Y. Babichenko, R. Peretz, P. Sousi and P. Winkler, and on earlier work by Adler et al (2003).) Add to calendar: 

Add to calendar :

Date/Time : 16 Aug 2018 14:00
Venue : S17 #04-05 SR2
Speaker : Dong Chaoping, Shanghai Normal University

Date/Time:16 Aug 2019 14:00 Venue: S17 #04-05 SR2 Speaker: Dong Chaoping, Shanghai Normal University Dirac cohomology: Towards the classification We will recall necessary math ingredients related to the notion of Dirac cohomology, such as Parthasarathy’s Dirac operator inequality, PRV-component, spin module, the unitarily small convex hull, and the spin norm. Then we will discuss some recent progress towards the classification of irreducible unitary representations with non-vanishing Dirac cohomology. In particular, we will mention a few questions. Add to calendar: 

Add to calendar :