AID2E bi-weekly technical meeting
Meeting: October 30, 2025
Attendees: Amit Bashyal, Fang-Ying Tsai, Torre Wenaus, Meifeng Lin, Derek Anderson, Tanja Horn, Baptiste Fraisse, Connor Pecar, Cynthia Nunez, Cristiano Fanelli, Karthik Suresh, Hemalata Nayak
- Announcements
- AI4EIC-2025 had just finished.
- Wen had presented on behalf of AID2E.
- The PI exchange was expected to be held over Zoom.
- Paper planning
- The paper should focus on technical capabilities.
- Target topics included multiple runners for toy functions and closure tests, plus multiple ePIC workflows combined into one AID2E workflow.
- Amit raised MOBO vs MOGA scope for the paper.
- Cris noted MOGA could be included if it was not too much extra work.
- The group discussed a global design-parameter space and a global problem definition.
- Task list
- Each sub-project was asked to identify one objective and provide startup instructions.
- Timelines for writeups were set around Nov 4, 2025.
- PanDA integration support from BNL was planned for several workflows.
- The writeups should explain shared ePIC utilities, including generator handling, reconstruction flags, and detector-geometry modification.
- Karthik raised software-version standardization.
- Amit noted different
eic-shell versions are manageable.
- Karthik was also starting work on a combined workflow, initially using Tracker Optimization and BIC-MOBO.
- Verbal updates
- BNL:
- Amit: Wen completed functionality to record objective runtime.
- dRICH-MOBO was being rerun with updated functionality.
- Pi/K separation objectives were being reworked into more stable definitions.
- Fang was finalizing convergence studies and comparing concurrency settings.
- Duke:
- Connor was working with the dRICH simulation team to redo geometry and reconstruction studies.
- JLab:
- Derek reported a BIC-MOBO SlurmRunner issue and ongoing analysis-script work.
- AOB
- Cris shared the Overleaf.
- Possible journals included Computing Software and Big Science and Computer Physics Communications.
- Action items
- By the general meeting, groups should identify one objective for their optimization problem.
- Karthik planned to report on instructions for running single workflows and on the steering file for multiple workflows.