Hsu:

Congratulations for the nice slide. The style is very good. I just have a few suggestions in some slides.

p3. There are several ATLAS jargons could be avoided (e.g. GRL, PV, ECIDS, DL1r, ...). Not all technical details required to be listed for public talk. 

      Some critical items should be clarified, perhaps even another slide with some cartons. e.g. SS,  

p5. Another abbreviations should be defined before introduction in this slide, e.g. CR, SR

p6. Is HT part of the input to SM BDT?  It's nice to clarify this, and add a backup slide to explain what variables enter BDT.
  
      What is the motivation of bkg reweighting training, and benefit of doing so?

p7. Another slide with several un-defined terms, HF, Conv. B-only, ...
      For this short term, you don't have to list all details. I would suggest to pick one of them and explain a bit more details, but just claim dedicated CR methods are developed and control other samples.

p9. you can choose to move this to backup slide if you run out of time.