[Theory-seminars] Seminars next week

[Carlota Andres Casas]

Dear all,

Here is a reminder for next week's seminars:

Theory Seminars:

-Monday, January 7th, 1:00PM, Room L102
Neill Warrington (U. of Maryland), "Deforming Path Integrals: Application to the (2+1) Thirring Model"

Abstract:

Systems of fermions at finite density have complex Boltzmann weights which cause the integrand of the path integral to become highly oscillatory. As a result of these oscillations, standard Monte Carlo integration methods require exponential precision in the spacetime volume to compute observables with reasonable accuracy. However, deforming the path integration contour to well-chosen manifolds in the complex plane can alleviate the sign problem. I will describe a deformation procedure which tames the sign problem called the "sign optimized manifold method", then apply it to the (2+1)d Finite Density Thirring Model, which is a theory with a sufficiently bad sign problem that standard Monte Carlo methods fail.


BlueJeans connection: https://urldefense.proofpoint.com/v2/url?u=https-3A__bluejeans.com_321085255&d=DwIFAw&c=lz9TcOasaINaaC3U7FbMev2lsutwpI4--09aP8Lu18s&r=60nExCXZj9SCzny0kAZk-AjVzjr42OsgTZnvcFFHHk8&m=Ub48L0rrbsKlmkDsjJpidt8onZ36NHlu4wVZTxnQH0M&s=ezK8P9xia1wHdFg05gSxLGKs1JImQaKYhA20u28k95Y&e= 


-Wednesday, January 9th, 1:00PM, Room L102
Michelle Kuchera (Florida State University), "Machine Learning for event simulation and classification in the Active-Target Time Projection Chamber."

Abstract:
The Active-Target Time Projection Chamber (AT-TPC) is a high-efficiency detector used for low-energy nuclear reactions at the National Superconducting Cyclotron Laboratory. A week-long experiment using the AT-TPC collects over 10 terabytes of data. Challenges in analyzing data include simulating detector response to events, especially simulation of noisy events, and classification of reactions. Generative adversarial networks (GANs) are used to improve simulations and convolutional neural networks (CNNs) are found to classify events most successfully. Implementation methods and results for the GANs and CNNs will be presented in the context of a proton elastic scattering experiment, 46Ar(p,p), with discussion of broader applications in nuclear physics.

BlueJeans connection: https://urldefense.proofpoint.com/v2/url?u=https-3A__bluejeans.com_195615763&d=DwIFAw&c=lz9TcOasaINaaC3U7FbMev2lsutwpI4--09aP8Lu18s&r=60nExCXZj9SCzny0kAZk-AjVzjr42OsgTZnvcFFHHk8&m=Ub48L0rrbsKlmkDsjJpidt8onZ36NHlu4wVZTxnQH0M&s=qxKW2PoXzZ1B7iinSDTsue17gYOmWJ3uBLy_AXadWMU&e= 


Raza, Vincent, Carlota
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/theory-seminars/attachments/20190104/55aa8c0b/attachment-0001.html>