AIOP Jun. 7, 2024

From epsciwiki
Jump to navigation Jump to search

The current weekly meeting time is every other Monday at 13:00 US/Eastern

Connection Info:

You can connect using ZOOM Video conferencing (ID: 161 588 8669). (Click "Expand" to the right for details -->):

One tap mobile: US: +16692545252,,1602484178# or +16468287666,,1602484178#

Meeting URL: https://jlab-org.zoomgov.com/j/1615888669?pwd=MnhGVUhpeEd3NVpnTGxoZHJManNBdz09&from=addon
Meeting ID: 161 588 8669
Passcode: 561441

Join by Telephone
For higher quality, dial a number based on your current location.
Dial: +1 669 254 5252 US (San Jose)
+1 646 828 7666 US (New York)
+1 646 964 1167 US (US Spanish Line)
+1 551 285 1373 US (New Jersey)
+1 669 216 1590 US (San Jose)
+1 415 449 4000 US (US Spanish Line)
833 568 8864 US Toll-free
Meeting ID: 161 588 8669
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: 161 588 8669
Passcode: 561441

SIP: 1615888669@sip.zoomgov.com
Passcode: 561441

One tap mobile: US: +16692545252,,1601987443# or +16468287666,,1601987443#

Meeting URL: https://jlab-org.zoomgov.com/j/1615888669?pwd=MnhGVUhpeEd3NVpnTGxoZHJManNBdz09&from=addon
Meeting ID: 161 588 8669
Passcode: 561441

AIOP - Polarized Photon Source

Agenda:

  1. Previous Meeting
  2. Announcements
  3. Project Progress
  4. ML Activities
  5. AOT



Minutes

AIOP Special Friday Meeting Notes

We should decide on a single framework for both projects

Armen has made some nice slides over this repo: SciOptControl Toolkit

Patrick’s questions:

How do we handle using a sparse data set in the offline/online environments?

Malachi — if we have a physical model, we can calibrate with historical data, then we can create the surrogate that we can wrap in openAI gym

Malachi suggests a white board meeting which is a good idea

We can try toy problems with different environments in openAI gymnasium

DS group really needs dataset summaries for each run period

David — there is a natural time scale for this, every measurement we take of the polarization is not necessarily incredibly accurate. Can we use information from more than 1 time step back? Do you have to give it a characteristic time scale or can it figure that out in the training process? Armen had an answer for this I just don't have what it was -- I think it is configurable though

Action Items

  • Placeholder

Milestones!

  • Identify relevant parameters
  • Identify nudge events
  • Shapley analysis!