sPHENIX discussion of physics

["Huang, Jin" <jhuang AT bnl.gov>]

Dear Cameron

 

Nice slides!

 

My main concern is that the ML part is delicate as presented on behalf of the collaboration. For example:

• Slide 14 states “Order of magnitude increase in HF statistics over triggered mode without ML tagging”, which draws questions when discussed together with the earlier pp projection plots in slides 9-13. The projection in BUP22 assumed 10% streaming recording, which is two to three orders of magnitude improvement on top of trigger mode, and already far surpassing the ML claim.
• “Predicted 75% background rejection by adding ML selection”: is a weak background rejection for HF measurements and the result trigger rate and purity was not discussed.

Therefore, it appears to me that instead of putting it as a new physics-ready capability, it is best to put the ML part as a demonstration run in Run24 with the aim of uses in EIC

 

Minor comments:

• Slide 2: good to add ZDCs
• Slide 6 These are very technical plots and therefore the text could be more descriptive to capture the main message?

• E.g. “5um hit resolution on MVTX” , “Good time resolution for INTT to resolve beam crossing”, “MVTX shows low noise <10^-8”

Slide 10 plot looks low resolution? Slide 11: “Full 𝐷0 → 𝐾𝜋 candidates are reconstructed, tracks then swapped with 𝐷0” sounds unclear. I assume it meant for “then swapped for mis-PID assumption.”?Between slide 12-13, we may need a backup slide on b-jet purity-efficiency projection.

 

Hope this is helpful.

 

Cheers

 

Jin

 

 

______________________________

 

Jin HUANG

 

Physicist, Ph.D.

Brookhaven National Laboratory

Physics Department, Bldg 510 C

Upton, NY 11973-5000

 

Office: 631-344-5898

Cell:   757-604-9946

______________________________

 

From: sPHENIX-physics-l <sphenix-physics-l-bounces AT lists.bnl.gov> On Behalf Of Dean, Cameron
Sent: Wednesday, August 23, 2023 9:37 AM
To: Perepelitsa, Dennis via sPHENIX-l <sphenix-l AT lists.bnl.gov>
Cc: sphenix-physics-l AT lists.bnl.gov; Antonio Silva via sPHENIX-HF-jets-l <sphenix-hf-jets-l AT lists.bnl.gov>
Subject: [Sphenix-physics-l] QM23 heavy flavor talk

 

Dear collaborators,

I will be giving the heavy flavor overview talk at Quark Matter. I’ve uploaded my slides to Invenio and they can be found here

My talk is 15 + 5. For context, the talk is scheduled to be given the day after Ed’s overview talk, 1 hour after Tim’s jet talk but before the poster session. The title of the talk that the organizers wrote down is “Heavy Flavor machine learning algorithms for Fast Data Processing in sPHENIX” but I have it titled as “Heavy flavor prospects at sPHENIX”. I also have 2 alternate slides at the start of my backup in case people think they’re better than what I have in the main body.

Please send me any comments and suggestions so I can best represent the awesome work of the collaboration.

Cheers,
Cameron