## Jing Jin

### Contact Information

## Dhamma Kimpara

### Contact Information

Improv comedy with a science twist!

Scientists, improvisation artists, and the audience combine to show the lighter side of science and life in the lab through short improv games and sketches.

Featuring Georgia Tech mathematician Lew Lefton.

#### Event Details

**Date/Time:**

Taste of Science is one of many Georgia Tech events at the 2018 Atlanta Science Festival.

Featuring live demonstrations, food samples, and fascinating facts that tie science, culture, and food together, the Taste of Science is sure to satisfy your appetite for learning.

Hosted by College of Sciences’ Ed Greco, Michael Evans, Jennifer Leavey, Enid Steinbart, and their students in students in the STEMcomm VIP class.

#### Event Details

**Date/Time:**

## Thibaud Alemany

### Contact Information

#### School of Mathematics' 2018 Stelson Lecture, by Jill C. Pipher, Brown University

How is it possible to send encrypted information across an insecure channel (like the internet) so that only the intended recipient can decode it, without sharing the secret key in advance?

In 1976, well before this question arose, a new mathematical theory of encryption (public-key cryptography) invented by Diffie and Hellman made digital commerce and finance possible. The technology advances of the last 20 years bring new and urgent problems, including the need to compute on encrypted data in the cloud and to have cryptography that can withstand the speed-ups of quantum computers.

In this lecture, Jill Pipher will discuss some of the history of cryptography and some of the latest ideas in "lattice" cryptography which, appear to be quantum resistant and efficient

RECEPTION FOLLOWS THE STELSON LECTURE

**About the Speaker**

Jill C. Pipher is Vice President for Research and Elisha Benjamin Andrews Professor of Mathematics at Brown University.

She received her Ph.D. from the University of California, Los Angeles. She is the president-elect of the American Mathematical Society and the first director of the Institute for Computational and Experimental Research.

Pipher taught at the University of Chicago before taking a position at Brown, where she served as chair of the Mathematics Department from 2005 to 2008.

Her research areas are harmonic analysis, Fourier analysis, partial differential equations, and cryptography. She has published more than 50 research articles and has coauthored a textbook on cryptography.

*Jill Pipher will deliver a colloquium on "Non-Smooth Boundary Value Problems" on Friday, March 2, 2018, 11 AM, in Room 006, Skiles Building, 686 Cherry St NW, Atlanta, GA 30313.*

**About the Stelson Lecture Series**

The series is made possible by an endowment created in 1988 by Thomas Stelson in honor of his father, Hugh Stelson. Thomas Stelson was a distinguished civil engineer who served Georgia Tech in high leadership positions from 1971 to 1990. Hugh Stelson was a mathematician who worked on problems related to interest rates, annuities, and numerical analysis. Lecturers invited for this series are first-rate mathematicians who are gifted speakers. Stelson lecturers give a public lecture for a general audience, as well as a colloquium for mathematicians and experts in related disciplines.

#### Event Details

**Date/Time:**

Jill C. Pipher, Vice President for Research and Elisha Benjamin Andrews Professor of Mathematics at Brown University will give the Stelson Lecture at Georgia Tech on March 1, 2018, 6:00-7:00 PM • Klaus Lecture Auditorium 1443, with a reception to follow at 7:00 PM.

See also the CoS story here:

http://www.cos.gatech.edu/hg/item/602067

Abstract for Stelson Lecture

How is it possible to send encrypted information across an insecure channel (like the internet) so that only the intended recipient can decode it, without sharing the secret key in advance? In 1976, well before this question arose, a new mathematical theory of encryption (public-key cryptography) was invented by Diffie and Hellman, which made digital commerce and finance possible. The technology advances of the last twenty years bring new and urgent problems, including the need to compute on encrypted data in the cloud and to have cryptography that can withstand the speed-ups of quantum computers. In this lecture, we will discuss some of the history of cryptography, as well as some of the latest ideas in "lattice" cryptography which appear to be quantum resistant and efficient

There will also be a Colloquium on Friday, March 2, 2018, at 11:00 AM in Skiles room 006.

Title: Non-smooth boundary value problems

The regularity properties of solutions to linear partial differential equations in domains depend on the structure of the equation, the degree of smoothness of the coefficients of the equation, and the boundary of the domain. Quantifying this dependence is a classical problem, and modern techniques can answer some of these questions with remarkable precision. For both physical and theoretical reasons, it is important to consider partial differential equations with non-smooth coefficients. We’ll discuss how some classical tools in harmonic and complex analysis have played a central role in answering questions in this subject at the interface of harmonic analysis and partial differential equations.

About the speaker:

Jill Pipher is the Elisha Benjamin Andrews Professor Mathematics at Brown University. She received her Ph.D. from the University of California, Los Angeles. She is the president-elect of the American Mathematical Society and she was the first director of the Institute for Computational and Experimental Research. She taught at the University of Chicago before taking a position at Brown, where she served as chair of the Mathematics Department from 2005 to 2008. Her work has been in harmonic analysis, Fourier analysis, partial differential equations, and cryptography. She has published more than 50 research articles and has coauthored a textbook on cryptography.

#### Event Details

**Date/Time:**

The workshop is a part of GTMAP activities of **the thematic semester on Control (Spring 2018)**.

This is a half day workhop on a panorama of mathematical questions in Dynamics and Control, presented by some people in the school of Mathematics and Physics.

2-2:30 Howie Weiss on "**Collateral sensitivity of antibiotic-resistant microbes: Modeling insights informing in-vitro studies**"

2:30-3 Predrag Cvitanovic on "**A spatiotemporal theory of turbulence : computational approaches**"

3-3:30 Luca Dieci on "**Periodic orbits of planar discontinuous systems under discretization**"

3:30- 4 BREAK

4-4:30 Federico Bonetto on "**Nash Equilibria in Decentralized Economies with Fiat Money**"

4:30-5 Haomin Zhou on "**A Brief Introduction to Optimal Transport on Finite Graphs**"

**Prof. Howie Weiss (GT Math, Emory Biology and Public Health)**

**Title:** Collateral sensitivity of antibiotic-resistant microbes: Modeling insights informing in-vitro studies

**Abstract:** Antibiotics have greatly reduced the morbidity and mortality due to infectious diseases. Antibiotic resistance constitutes a significant threat to human health. One strategy to help combat resistance is to find novel ways to use existing drugs, even those that display high rates of resistance. For some species, pairs of antibiotics have been identified for which evolution of resistance to drug A increases sensitivity to drug B and visa versa. These research groups have proposed cycling such pairs to treat infections, and similar treatment strategies are being investigated to treat various cancer forms as well.

While an exciting treatment prospect, no cycling experiments have yet been performed with consideration of pharmacokinetics (PK) and pharmacodynamics (PD). To test the plausibility of this scheme and search for ways to optimize it, we create a mathematical model with explicit PK-PD considerations. We study several possible treatment protocols using pairs of collaterally sensitive antibiotics, and investigate the speed of ascent of multiply resistant mutants.

**Prof. Predrag Cvitanovic (GT Physics)**

**Title:** A spatiotemporal theory of turbulence : computational approaches

**Abstract:** The recurrent flows observed in moderate Reynolds number turbulence are shaped by close passes to unstable invariant solutions of Navier-Stokes equations. While in recent years many such solutions been computed, so far all have been confined to small computational domains.

Pipe, channel and plane flows, however, are flows on infinite spatial domains. If the Navier-Stokes equations are recast as a space-time theory, with both space and time taken to infinity, the traditional Direct Numerical Simulation codes have to be abandoned. In this theory there is no time, there is only a repertoire of admissible spatiotemporal patterns. To determine these, radically different kinds of codes will have to be written, with space and time treated on equal footing: some of the possible approaches will be sketched.

**Prof. Luca Dieci (GT Math)**

**Title:** Periodic orbits of planar discontinuous systems under discretization

**Abstract:** We consider a piecewise smooth planar system having an attracting periodic orbit of transversal type. For this, we look at the discrete system resulting from a Euler discretization with fixed stepsize. We show that, in general, the resulting discrete dynamical system does not possess an invariant curve, unlike what happens for smooth problems. Still, we show that the discrete trajectories are forced to remain inside a band, whose width is proportional to the stepsize. We further show that by forcing the numerical solution to step exactly on the discontinuity line, then there is a discrete periodic solution near the one of the original problem. Finally, we consider what happens to the Euler discretization of the regularized system, and give evidence that the discrete solution undergoes a sequence of period doubling bifurcations with respect to the regularization parameter. Joint work with Timo Eirola and Cinzia Elia.

**Prof. Federico Bonetto (GT Math)****Title**: Nash Equilibria in Decentralized Economies with Fiat Money

**Abstract**: We study the dynamics of a simple economy made by agents that can exchange goods and fiat money. The model was initially introduced '97 by Kiyotaki and Wright to discuss the origin of fiat money and its relation with commodity money. Notwithstanding its simplicity, the dynamics it generate is rather challenging. I'll try to explain the issues involved and possible solutions.

**Prof. Haomin Zhou (GT Math)****Title**: A Brief Introduction to Optimal Transport on Finite Graphs

**Abstract**: Optimal transport theory in continuous space has been extensively studied in

the past few decades. In this talk, I will present similar matters on discrete

spaces. Various recent developments related to free energy,

Fokker-Planck equations, as well as Wasserstein distance on graphs will be discussed,

some of them are rather surprising. This presentation is based on several joint

papers with Shui-Nee Chow (Georgia Tech), Wen Huang (USTC), Wuchen Li (UCLA),

Yao Li (U. Mass), Haoyan Zhai (Georgia Tech).

#### Event Details

**Date/Time:**

This is a part of the GT MAP activities on Control. GT MAP is a place for research discussion and collaboration. We welcome participation of any researcher interested in discussing his/her project and exchange ideas with Mathematicians.

There will be light refreshments through out the event. This seminar will be held in** Skiles 006** and refreshments at Skiles Atrium.

A couple of members of Prof. Pandarinath's group will present their research

3:00 PM - 3:45PM **Prof. Pandarinath **will give a talk on "**Unsupervised discovery of ensemble dynamics in the brain using deep learning techniques**."

3:45PM -- 4:00PM Break with Discussions

4:00PM - 4:25PM Second talk given by **Lahiru Wimalasena **(Graduate student, Wallace H. Coulter Department of Biomedical Engineering, Emory University and Georgia Tech).

4:25PM - 5PM Discussion of open problems stemming from the presentations.

**Title:** Unsupervised discovery of ensemble dynamics in the brain using deep learning techniques

**Abstract: **

Since its inception, neuroscience has largely focused on the neuron as the functional unit of the nervous system. However, recent evidence demonstrates that populations of neurons within a brain area collectively show emergent functional properties ("dynamics"), properties that are not apparent at the level of individual neurons. These emergent dynamics likely serve as the brain’s fundamental computational mechanism. This shift compels neuroscientists to characterize emergent properties – that is, interactions between neurons – to understand computation in brain networks. Yet this introduces a daunting challenge – with millions of neurons in any given brain area, characterizing interactions within an area, and further, between brain areas, rapidly becomes intractable.

I will demonstrate a novel unsupervised tool, Latent Factor Analysis via Dynamical Systems ("LFADS"), that can accurately and succinctly capture the emergent dynamics of large neural populations from limited sampling. LFADS is based around deep learning architectures (variational sequential auto-encoders), and builds a model of an observed neural population's dynamics using a nonlinear dynamical system (a recurrent neural network).

When applied to neuronal ensemble recordings (~200 neurons) from macaque primary motor cortex (M1), we find that modeling population dynamics yields accurate estimates of the state of M1, as well as accurate predictions of the animal's motor behavior, on millisecond timescales. I will also demonstrate how our approach allows us to infer perturbations to the dynamical system (i.e., unobserved inputs to the neural population), and further allows us to leverage population recordings across long timescales (months) to build more accurate models of M1's dynamics.

This approach demonstrates the power of deep learning tools to model nonlinear dynamical systems and infer accurate estimates of the states of large biological networks. In addition, we will discuss future directions, where we aim to pry open the "black box" of the trained recurrent neural networks, in order to understand the computations being performed by the modeled neural populations.

pre-print available: lfads.github.io

#### Event Details

**Date/Time:**

This is a part of the GT MAP activities on Control. GT MAP is a place for research discussion and collaboration. We welcome participation of any researcher interested in discussing his/her project and exchange ideas with Mathematicians.

There will be light refreshments through out the event. This seminar will be held in** Skiles 006** and refreshments at Skiles Atrium.

A couple of members of **Prof. Theodorou**'s group will present their research

3:00 PM - 3:45PM** Prof. Theodorou **will give a talk on " **The science of autonomy: A "happy" symbiosis between learning, control and physics**."

3:45PM -- 4:00PM Break with Discussions

4:00PM - 4:25PM another talk.

4:25PM - 5PM Discussion of open problems stemming from the presentations.

**Title:** The science of autonomy: A "happy" symbiosis between learning, control and physics.

**Abstract: ** In this talk I will present an information theoretic approach to stochastic optimal control and inference that has advantages over classical methodologies and theories for decision making under uncertainty. The main idea is that there are certain connections between optimality principles in control and information theoretic inequalities in statistical physics that allow us to solve hard decision making problems in robotics, autonomous systems and beyond. There are essentially two different points of view of the same "thing" and these two different points of view overlap for a fairly general class of dynamical systems that undergo stochastic effects. I will also present a holistic view of autonomy that collapses planning, perception and control into one computational engine, and ask questions such as how organization and structure relates to computation and performance. The last part of my talk includes computational frameworks for uncertainty representation and suggests ways to incorporate these representations within decision making and control.

#### Event Details

**Date/Time:**