Difference between revisions of "AIEC Weekly Meeting May 5, 2022"
(Created page with "=== Connection Info: === <div class="toccolours mw-collapsible mw-collapsed"> You can connect using [https://jlab-org.zoomgov.com/j/1601987443?pwd=bUVPV3puTHJPU0Fub2Q2Z1dsV0Jj...") |
|||
Line 31: | Line 31: | ||
=== Agenda: === | === Agenda: === | ||
# [[AIEC Weekly Meeting Apr. 28, 2022| Previous Meeting]] | # [[AIEC Weekly Meeting Apr. 28, 2022| Previous Meeting]] | ||
+ | #: | ||
+ | # UVA Capstone ([https://docs.google.com/document/d/1lbb89vG8r1_EpXiBY0NXIRA9fGKJ2X06kGNZNP2alJ0/edit?usp=sharing 2021 application]) | ||
#: | #: | ||
# Activities: | # Activities: |
Latest revision as of 18:12, 5 May 2022
Connection Info:
You can connect using ZOOM Video conferencing (ID: 1601987443). (Click "Expand" to the right for details -->):
One tap mobile: US: +16692545252,,1601987443# or +16468287666,,1601987443#
Meeting URL: https://jlab-org.zoomgov.com/j/1601987443?pwd=bUVPV3puTHJPU0Fub2Q2Z1dsV0JjQT09&from=addon
Meeting ID: 160 198 7443
Passcode: 607906
Join by Telephone
For higher quality, dial a number based on your current location.
Dial:
US: +1 669 254 5252 or +1 646 828 7666 or +1 669 216 1590 or +1 551 285 1373 or 833 568 8864 (Toll Free)
Meeting ID: 160 198 7443
International numbers
Join from an H.323/SIP room system
H.323: 161.199.138.10 (US West)
161.199.136.10 (US East)
Meeting ID: 160 198 7443
Passcode: 607906
SIP: 1601987443@sip.zoomgov.com
Passcode: 607906
Agenda:
- Previous Meeting
- UVA Capstone (2021 application)
- Activities:
- BCAL baseline (slides)
- Cosmics run analysis
- Uncertainty Quantification
- CPP preparations
- FDC calibration
- Next steps
- CLAS12 ECAL PMT gain
- GlueX TOF
- Talks/Presentations
- Publications:
- AI4EIC Proceedings accepted AI@EIC Overleaf
- ACAT Proceedings submitted ACAT Overleaf
- NIM paper in progress NIM Overleaf
Important Dates
- 2022-06-07 : CPP projected start date (Experiment Schedule)
Notes
- Evaluate uncertainty
- 1, 2, 3 sigma tests on holdout from 2020 and 2021 data
- check uncertainty for a slice of data in the high uncertainty zone to see if it is stable or gradually increasing.
- DECISION: how to use uncertainty?
- When uncertainty over a threshold project to "closest" point on a spherical map of inputs to uncertainty, and use the GCF/HV setting for the closest point.
- DECISION: what is the threshold?
- 0.004
- DECISION: Which model to use in CPP
- GPR with learned prior, trained on 2020 and 2021 data.