Date/Time : 20 Jan 2020 15:00
Venue : S17 #04-06 SR1
Speaker : Sohail Bahmani, Georgia Institute of Technology

Date/Time:20 Jan 2020 15:00 Venue: S17 #04-06 SR1 Speaker: Sohail Bahmani, Georgia Institute of Technology Convex Programming for Statistical Estimation: A New Perspective A critical aspect of statistical procedures in many engineering and scientific applications is their computational efficiency. Convex programming provides a versatile framework to design computationally tractable algorithms for many statistical tasks such as regression, classification, and variable selection. The estimation procedures based on convex programming are traditionally designed independent of the statistical model. We will discuss new approaches that use the statistical model to design estimators with convex formulations. These approaches can be applied in a broader set of problems and be computationally favourable. In particular, we consider a broad class of parametric regression problems where the signal is observed through random nonlinear functions with a difference of convex (DC) form. This general statistical model includes familiar special cases such as phase retrieval/quadratic regression and blind deconvolution/bilinear regression. Given the DC decomposition of the observation functions as well as an approximate solution, we formulate a convex program as an estimator that operates in the natural space of the signal. In the mentioned special cases, the proposed approach is computationally superior to the methods based on semidefinite/sum-of-squares relaxation—tailored for polynomial observation functions—and can compete with the non-convex methods. Under some regularity assumptions, the proposed convex estimators achieve a desirable sample complexity. Add to calendar: 

Add to calendar :

Date/Time : 17 Jan 2020 14:00
Venue : S16 #05-98
Speaker : Xinyi Li , Peiking University

Date/Time:17 Jan 2020 14:00 Venue: S16 #05-98 Speaker: Xinyi Li, Peiking University Recent Progress on Loop-erased Random Walk In this mini-course, we will give a brief review on the model of Loop-erased Random Walk (LERW) and its relationship with other models from statistical physics, and discuss some of the recent progress on this model, especially in three dimensions. Add to calendar: 

Add to calendar :

Date/Time : 16 Jan 2020 14:00
Venue : S16 #05-98
Speaker : Xinyi Li , Peiking University

Date/Time:16 Jan 2020 14:00 Venue: S16 #05-98 Speaker: Xinyi Li, Peiking University Recent Progress on Loop-erased Random Walk In this mini-course, we will give a brief review on the model of Loop-erased Random Walk (LERW) and its relationship with other models from statistical physics, and discuss some of the recent progress on this model, especially in three dimensions. Add to calendar: 

Add to calendar :

Date/Time : 15 Jan 2020 14:00
Venue : S16 #05-98
Speaker : Xinyyi Li , Peiking University

Date/Time:15 Jan 2020 14:00 Venue: S16 #05-98 Speaker: Xinyyi Li, Peiking University Recent Progress on Loop-erased Random Walk In this mini-course, we will give a brief review on the model of Loop-erased Random Walk (LERW) and its relationship with other models from statistical physics, and discuss some of the recent progress on this model, especially in three dimensions. Add to calendar: 

Add to calendar :

Date/Time : 15 Jan 2020 14:00
Venue : S17 #04-06 SR1
Speaker : Paul Nelson, ETH Zurich

We will discuss how to study the cubic moment of any family of automorphic L-functions on PGL(2) using regularized diagonal periods of Eisenstein series, following a strategy suggested by Michel–Venkatesh. Applications include generalizations to the setting of number fields of some results of Conrey–Iwaniec and Petrow–Young, improved estimates for representation numbers of ternary quadratic forms over number fields, and improvements to the prime geodesic theorem on arithmetic hyperbolic 3-folds.

Add to calendar :

Date/Time : 15 Jan 2020 15:00
Venue : S17 #06-11
Speaker : Georgios Moschidis, UC Berkeley

Date/Time:15 Jan 2020 15:00 Venue: S17 #06-11 Speaker: Georgios Moschidis, UC Berkeley A proof of the instability of AdS spacetime for the Einstein–scalar field system The AdS instability conjecture is a conjecture about the initial-boundary value problem in general relativity. It states that there exist arbitrarily small perturbations to the initial data of AdS spacetime which, under evolution by the vacuum Einstein equations with reflecting boundary conditions on conformal infinity, lead to the formation of black holes after a sufficiently long time. In this talk, I will present a rigorous proof of the AdS instability conjecture in the setting of the spherically symmetric Einstein–scalar field system. The construction of the unstable family of initial data will require working in a low regularity setting, carefully designing a family of initial configurations of localised matter beams and estimating the exchange of energy taking place between interacting beams over a long period of times, as well as estimating the decoherence rate of the beams. Add to calendar: 

Add to calendar :

Date/Time : 15 Jan 2020 15:30
Venue : S17 #04-04 SR3
Speaker : Qi Zhi, Zhejiang University

Date/Time:15 Jan 2020 15:30 Venue: S17 #04-04 SR3 Speaker: Qi Zhi, Zhejiang University A Bessel delta-method and its applications In this talk, I will introduce a new $\delta$-method by a certain Bessel integral and its applications in the theory of exponential sums for $GL(2)$. In particular, it gives a simple proof of the Weyl-type subconvex bound for modular form $L$-functions in the $t$-aspect. This is joint work with Aggarwal, Holowinksy and Lin. Add to calendar: 

Add to calendar :

Date/Time : 15 Jan 2020 17:00
Venue : S17 #04-06 SR1
Speaker : Frank Stephan, National University of Singapore

Date/Time:15 Jan 2020 17:00 Venue: S17 #04-06 SR1 Speaker: Frank Stephan, National University of Singapore Measure and Conquer for Max Hamming Distance XSAT This talk gives an overview of the joint work of the speaker’s ecent work on the Hamming XSAT problem. This problem asks for an algorithm to determine which, given an XSAT instance, determines the maximum Hamming distance between two solutions of this instance. The problem has been studied by Dahloef in 2005 in an ISAAC paper who provided an O(1.83848^n) algorithm for this problem. Later Fu, Zhu and Yin presented at JFCST 2012 an O(1.6760^n) algorithm for the related Max Hamming X3SAT problem where all clauses have at most three literals. The current paper provides for the general Max Hamming XSAT problem an O(1.4983^n) algorithm which applies also the technique “Measure and Conquer” in order to prove a better bound than the algorithm would give otherwise. Furthermore, the algorithm does not only provide the maximum Hamming distance of two solutions of the instance, but instead for each k between 0 and n the number of pairs of solutions which have Hamming distance k. For the special case Max Hamming Distant X3SAT, Hoi, Jain and Stephan have the bound O(1.3298^n) at the conference FSTTCS 2019. Add to calendar: 

Add to calendar :

Date/Time : 13 Jan 2020 15:00
Venue : S17 #04-06 SR1
Speaker : Zhu Wei, Duke University

Date/Time:13 Jan 2020 15:00 Venue: S17 #04-06 SR1 Speaker: Zhu Wei, Duke University Applied differential geometry and harmonic analysis in deep learning regularization With the explosive production of digital data and information, data-driven methods, deep neural networks (DNNs) in particular, have revolutionized machine learning and scientific computing by gradually outperforming traditional hand-craft model-based algorithms. While DNNs have proved very successful when large training sets are available, they typically have two shortcomings: First, when the training data are scarce, DNNs tend to suffer from overfitting. Second, the generalization ability of over parameterized DNNs still remains a mystery despite many recent efforts. In this talk, I will discuss two recent works to inject” the “modelling” flavour back into deep learning to improve the generalization performance and interpretability of DNNs. This is accomplished by deep learning regularization through applied differential geometry and harmonic analysis. In the first part of the talk, I will explain how to improve the regularity of the DNN representation by imposing a “smoothness” inductive bias over the DNN model. This is achieved by solving a variation problem with a low-dimensionality constraint on the data-feature concatenation manifold. In the second part, I will discuss how to impose scale-equivariance in network representation by conducting joint convolutions across the space and the scaling group. The stability of the equivalent representation to nuisance input deformation is also proved under mild assumptions on the Fourier-Bessel norm of filter expansion coefficients. Add to calendar: 

Add to calendar :