Institute for Nuclear and Particle Astrophysics (INPA) at LBNL

The INPA Seminar weekly talks are on Fridays, starting at 12:00 pm, unless informed otherwise. The seminar talk starts with a brief presentation of the weekly scientific news. Typically, the talks conclude by 1:00 pm. The seminars are held in the Sessler Conference Room,  located in Bldg. 50A- 5132.

The committee members are:

The seminar schedule for the Institute for Nuclear and Particle Astrophysics (INPA) is tentative and becomes final a few days before the Friday talk.

Please send all suggestions for future INPA talks and speakers to the INPA Committee.

To be added to the INPA News Mailing List, please contact Erica Hall.

Loading Events

« All Events

  • This event has passed.

[Special] Speaker: Mark Anderson(Queen’s University) Title: Advancing nuclear and particle physics with deep learning: from event reconstruction to signal denoising

February 1 @ 12:00 pm - 1:00 pm


Date:  Thursday, February 1, 2024

Time: 12:00 PM – 1:00 PM

Location: INPA Conference Room–5026 [Hybrid and In-Person]

Speaker: Mark Anderson (Queen’s University)

Title: Advancing nuclear and particle physics with deep learning: from event reconstruction to signal denoising

Abstract: In this talk, I will present my research on advancing nuclear and particle physics using deep learning. The first portion of the presentation will focus on the SNO+ experiment, a multi-purpose neutrino experiment with the primary goal of searching for neutrinoless double-beta decay. In particular, my work involves developing deep learning-based techniques for event reconstruction of particle interactions in the detector. Currently, SNO+ uses a chain of algorithms involving maximum likelihood estimation. I will demonstrate that my independent, deep learning approach can be used to improve the reconstruction accuracy, while also offering substantial gains in speed. This has implications on volume fiducialization and background rejection.
In the second part of my talk, I will present studies on a deep convolutional autoencoder designed to remove electronic noise from a p-type point contact high purity germanium detector. With their intrinsic purity and excellent energy resolutions, these detectors are suitable for a variety of rare event searches such as neutrinoless double-beta decay, dark matter candidates, and other exotic physics. However, noise from the readout electronics can make identifying events of interest more challenging. I will highlight several studies that show that detector performance can be improved when signals are denoised with the autoencoder, leading to better physics outcomes.
While both of these projects are applied to data from specific detector technologies, the methods I have developed are broadly applicable. The groups I work with have begun exploring new applications of my research with promising results. At the end of the talk, I will discuss some specific extensions and provide an outlook for the future.

Join Zoom Meeting

Meeting ID: 950 1669 6011

Passcode: 247722


February 1
12:00 pm - 1:00 pm

INPA guests from campus can now come to the lab early on Fridays. The INPA Common Room (50-5026) is reserved for our guests from 10:00 a.m. to 12:00 noon and from 1 p.m. to 5 p.m. Note that the seminars are now held in 50A-5132 to accommodate a more significant number of attendees.

CPTea Series (also known as INPA Tea Series)

The Physics Division CPTea Series invites you to an In-Person Tea Series 1st Friday of every month at 3:30 pm INPA Conference Room 50-5026.

Everyone is welcome to attend the open forum. Tea and light refreshments will be served.


INPA Common Room (50-5026)
3:30 pm

Access to the Lab

For a shuttle pass, please email Erica Hall. The pass is only valid for the day of the seminar.

Erica Hall