Probing the flavor structure of New Physics models with precision observables

Smolkovič, Aleks (avtor), Košnik, Nejc (mentor)

The Standard Model of particle physics is reviewed with the emphasis on the flavor aspects of the theory, together with its many shortcomings. The enhancement of CP asymmetries in $B \to K \mu \mu$ due to the effects of narrow charmonia is studied and proposed as a viable probe of the CP nature of the physics beyond the Standard Model, which might be entering in $b \to s \mu \mu$ transitions. Such direct CP asymmetries offer probes that are highly complementary to the observables sensitive to lepton flavor universality violation, which show promising experimental deviations from the Standard Model predictions. Optimized probes of the CP nature of the top quark Yukawa coupling are studied both in single top quark as well as top-anti top quark pair associated production with the Higgs boson at proton-proton colliders. Estimated bounds from using such probes are given for the High Luminosity and High Energy Large Hadron Collider, as well as the hadronic Future Circular Collider. A connection between flavor anomalies and dark matter is attempted in the context of leptoquark models and a real scalar singlet dark matter candidate. A high-mass dark matter regime is pointed out as viable by considering constraints from relic abundance, direct, indirect, and collider searches, as well as the scalar potential stability. The combination of these constraints offers a novel bound on the leptoquark mass. An unsupervised machine learning technique based on variational autoencoders is presented as a powerful tool that could help at characterizing difficult signal signatures at colliders. Such an approach offers a highly competitive classification performance, and shows promising results for the characterization of signal properties by using the compressed representation of data in the latent space.

• Jezik: Angleški jezik
• Ključne besede: beyond the Standard Model, flavor physics, CP violation, optimized observables, dark matter, leptoquark, machine learning in particle physics
• Vrsta gradiva: Doktorsko delo/naloga
• Tipologija: 2.08 - Doktorska disertacija
• Organizacija: FMF - Fakulteta za matematiko in fiziko
• Leto izida: 2021
• PID: 20.500.12556/RUL-131039
• COBISS.SI-ID: 83717635
• Datum objave v RUL: 22.09.2021

Sekundarni jezik

• Jezik: Slovenski jezik
• Naslov: Odkrivanje strukture okusa v modelih nove fizike z natančnimi opazljivkami

Izvleček

V delu sprva povzamemo Standardni model osnovnih delcev s poudarkom na fiziki okusa osnovnih delcev, ter izpostavimo njegove številne pomanjkljivosti. Preučujemo povišanje vrednosti asimetrij CP v razpadu $B \to K \mu \mu$ kot posledico efektov ozkih čarmonijevih resonanc. Takšne direktne asimetrije CP bi lahko pripomogle k določanju lastnosti CP fizike onkraj Standardnega Modela, ki morda nastopa v prehodih $b \to s \mu \mu$, kot na to namigujejo opazljivke, občutljive na kršitev univerzalnosti leptonskega okusa. V produkciji kvarka top, kot tudi para kvarkov top, skupaj s Higgsovim bozonom na protonskih trkalnikih, predlagamo optimizirane opazljivke, občutljive na naravo CP Yukawa sklopitve kvarka top. Podamo oceno senzitivnostnih meja, ki bi jih lahko z uporabo predlaganih opazljivk dosegli hadronski trkalniki v prihodnosti. Skušamo povezati anomalije okusa s problemom temne snovi v kontekstu modelov s skalarnim leptokvarkom in realnim skalarnim singletom kot kandidatom za temno snov. Izpostavimo režim visokih mas temne snovi, v katerem so modeli konsistentni s trenutnimi omejitvami iz meritev preostale gostote temne snovi v vesolju, direktnih in indirektnih iskanj, iskanj na trkalnikih, ter s stabilnostjo skalarnega potenciala. Kombinacija teh omejitev ponuja novo omejitev na dovoljeno maso leptokvarka. Predstavimo variacijski samokodirnik, ki spada pod tehnike nenadzorovanega strojnega učenja, kot orodje, ki bi lahko pomagalo pri iskanju in opisu lastnosti redkih signalov na trkalnikih. Takšen pristop ponuja konkurenčno klasifikacijsko zmogljivost in kaže obetavne rezultate za karakterizacijo lastnosti signala z uporabo stisnjene upodobitve podatkov v latentnem prostoru.

• Ključne besede: fizika onkraj Standardnega Modela, fizika okusa, kršitev simetrije CP, optimizirane opazljivke, temna snov, leptokvark, strojno učenje v fiziki osnovnih delcev