sheride / ttbarzp-ml

Machine learning with p p > t tbar Z' event data

ttbarzp-ml

A generic beyond the Standard Model (BSM) theory involves the introduction a Z' particle, a neutral heavy U(1) gauge boson. Such a particle could, in particular, explain the B-meson anomalies [1] or the g-2 muon anomaly [2]; however, one must also consider why such a particle hasn't yet been observed at the LHC. We endeavor to explain the experimental anomalies along with our present discovery failures by considering a Z' which couples strongest to the third generation quarks.

We embark on a collider phenomenology program investigating the p p > t t~ Z' channel, wherein the Z' may decay to, e.g., b b~, t t~, tau tau~, or mu mu~. Likewise, each top may decay hadronically (t > W b > j j b) or leptonically (t > W b > l~ nu b).

The difficulty that arises in discriminating signal from background (which include, e.g., p p > t t~ h and p p > t t~ t t~) motivates the use of machine learning methodologies. This github repository documents code developed and employed to study this binary classification problem, focusing on the Z' > b b~ semihadronic-semileptonic channel.

[1] LHCb collaboration, et al. "Test of lepton universality in beauty-quark decays." (2021).

[2] Abi, B. et al. "Measurement of the Positive Muon Anomalous Magnetic Moment to 0.46 ppm". Phys. Rev. Lett. 126. (2021): 141801.