For 30+ years, particle physicists' predictions for where breakthrough discoveries are waiting have been wrong. False promises based on such flawed predictions erode trust in physics - and science in general. Particle physicists should fix this problem.https://www.nytimes.com/2019/01/23/opinion/particle-physics-large-hadron-collider.html …
-
3) Computation and infrastructure is treated as a stepchild in physics to a big detriment. While so much effort is spent ensuring every part of accelerator is calibrated, on data side, it feels like the wild west.
4) @skdh talks about spurious discoveries due to noise. Statistical techniques of multiple testing and false discovery rates deal with this. Although not completely bullet proof (as seen in genome wide studies fiasco) can be a start. Did papers use such techniques?
@SaltyBurger
-
-
False HEP discovery claim I remember: Dama (but the effect is true. The interpretation was wrong) and superluminar neutrinos (a cable issue). None at LHC just because the threshold is high (5 sigmas after trial factor). See 750 GeV diphoton (never claimed a discovery)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
In any case, there is no paper I can remember claiming a false discovery at colliders published on a HEP journal. Not sure which sources you have. Ask your HEP colleagues there (
@MariaSpiropulu ) for more accurate first-hand information than a random claim on newspapersThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
While your point on methods is valuable, as a matter of fact the requirement for a discovery claim is so high that at that point you are sure that what you see is not noise. A discovery that depends on the statistical method will not be taken seriously.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
On a more general ground, I would say that what we need is injecting know-how from
#datascience to HEP and you have a point there. Some of us workingon it (see eg https://docs.google.com/document/d/1FtoFRaHMZ1VFUqKHzzus_PCyyriJMSgj8S5uDPo_Y7M …)Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Among all discussion about failures&success in HEP (as an optimist I assumes we ALL lead those enthusiastically because we deeply care about science), there's one thing exp. HEP can barely be criticised for: the deep caring for statistical relevance & caution of discovery claim.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.