Difference between revisions of "AIOP Aug. 12, 2024"

From epsciwiki
Jump to navigation Jump to search
(Created page with "The current weekly meeting time is every other Monday at 13:00 US/Eastern === Connection Info: === <div class="toccolours mw-collapsible mw-collapsed"> You can connect using...")
 
 
(One intermediate revision by the same user not shown)
Line 70: Line 70:
 
=== Minutes ===
 
=== Minutes ===
 
<!-- Attendees: Torri, David, Jiawei, Armen, Hovannes, Malachi -->
 
<!-- Attendees: Torri, David, Jiawei, Armen, Hovannes, Malachi -->
−
 
+
<ul>
−
 
+
  <li>Attendees: Thomas, David, Torri, Armen, Patrick, Jiawei, Naomi, Cristiano</li>
 +
  <li>Progress
 +
    <ul>
 +
      <li>Jiawei
 +
        <ul>
 +
          <li>Showed pair plots from all nudge sequences
 +
            <ul>
 +
              <li>By polarization orientation</li>
 +
            </ul>
 +
          </li>
 +
          <li>Naomi questioned why there is not just 1 beam spot
 +
            <ul>
 +
              <li>Beam wobble?
 +
                <ul>
 +
                  <li><a href="https://logbooks.jlab.org/entry/4140883">https://logbooks.jlab.org/entry/4140883</a></li>
 +
                  <li><a href="https://logbooks.jlab.org/entry/4141000">https://logbooks.jlab.org/entry/4141000</a></li>
 +
                </ul>
 +
              </li>
 +
            </ul>
 +
          </li>
 +
          <li>Thomas: can we get correlation with pitch and yaw together
 +
            <ul>
 +
              <li>C angle?</li>
 +
              <li>Hovannes Hidden offsets</li>
 +
              <li>Use deltas?</li>
 +
              <li>David speculates finding the offsets by plotting the edge to pitch, yaw
 +
                <ul>
 +
                  <li>Livingston an approximate…</li>
 +
                </ul>
 +
              </li>
 +
            </ul>
 +
          </li>
 +
          <li>Naomi suggests filtering based on beam spot
 +
            <ul>
 +
              <li>Pretty easy to do. Maybe reclassify nudge sequences based on beam stability</li>
 +
            </ul>
 +
          </li>
 +
        </ul>
 +
      </li>
 +
      <li>Armen
 +
        <ul>
 +
          <li>We have a non-differentiable situation due to complex if/then</li>
 +
          <li>Actor needs to have a Siamese part or something to mimic the conditional</li>
 +
          <li>Is this environment suitable?
 +
            <ul>
 +
              <li>Livingston equation sufficient?</li>
 +
            </ul>
 +
          </li>
 +
          <li>Patrick: We aren’t using roll yet. Using roll will give the orientation…may eliminate conditionals. Needs more thought and a talk with Hovannes</li>
 +
          <li>David worries including roll may lead to effective discontinuities
 +
            <ul>
 +
              <li>Maybe we can engineer something smoother</li>
 +
            </ul>
 +
          </li>
 +
          <li>Naomi: We know what the button does…
 +
            <ul>
 +
              <li>Maybe we teach it to push the button</li>
 +
            </ul>
 +
          </li>
 +
          <li>Thomas: Train to classify the button press or not press</li>
 +
          <li>Armen…descretize the control space; more classically RL</li>
 +
          <li>Armen: Can we descretize the environment currently</li>
 +
          <li>Discussion of potential issues with taking the other track</li>
 +
        </ul>
 +
      </li>
 +
    </ul>
 +
  </li>
 +
  <li>Coming to a consensus to take a step back
 +
    <ul>
 +
      <li>We are going to abstract the buttons out. Train it to be the shift taker first</li>
 +
    </ul>
 +
  </li>
 +
</ul>
  
 
=== Milestones! ===
 
=== Milestones! ===

Latest revision as of 16:43, 26 August 2024

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
    • Quarterly sent to DOE
  3. Project Progress
  4. PIER Activities
    • Middle School Data Science Hackathon Workshop
    • HUGS mini-Workshop - COMPLETED!
  5. AOT



Minutes

  • Attendees: Thomas, David, Torri, Armen, Patrick, Jiawei, Naomi, Cristiano
  • Progress
    • Jiawei
    • Armen
      • We have a non-differentiable situation due to complex if/then
      • Actor needs to have a Siamese part or something to mimic the conditional
      • Is this environment suitable?
        • Livingston equation sufficient?
      • Patrick: We aren’t using roll yet. Using roll will give the orientation…may eliminate conditionals. Needs more thought and a talk with Hovannes
      • David worries including roll may lead to effective discontinuities
        • Maybe we can engineer something smoother
      • Naomi: We know what the button does…
        • Maybe we teach it to push the button
      • Thomas: Train to classify the button press or not press
      • Armen…descretize the control space; more classically RL
      • Armen: Can we descretize the environment currently
      • Discussion of potential issues with taking the other track
  • Coming to a consensus to take a step back
    • We are going to abstract the buttons out. Train it to be the shift taker first

Milestones!

Retrieved from "https://wiki.jlab.org/epsciwiki/index.php?title=AIOP_Aug._12,_2024&oldid=5391"