Abstract: We present a new algorithm for anomaly detection called Anomaly Awareness. The algorithm learns about normal events while being made aware of the anomalies through a modification of the cost function. We show how this method works in different Particle Physics situations and in standard Computer Vision tasks. For example, we apply the method to images from a Fat Jet topology generated by Standard Model Top and QCD events, and test it against an array of new physics scenarios, including Higgs production with EFT effects and resonances decaying into two, three or four subjets. We find that the algorithm is effective identifying anomalies not seen before, and becomes robust as we make it aware of a varied-enough set of anomalies.

Abstract: Dark-matter particles that annihilate or decay can undergo complex sequences of processes, including strong and electromagnetic radiation, hadronisation, and hadron de-cays, before particles that are stable on astrophysical time scales are produced. Antiprotons produced in this way may leave footprints in experiments such as AMS-02. Several groups have reported an excess of events in the antiproton flux in the rigidity range of 10-20 GV. However, the theoretical modeling of baryon production is not straightforward and relies in part on phenomenological models in Monte Carlo event generators. In this work, we assess the impact of QCD uncertainties on the spectra of antiprotons from dark-matter annihila-tion. As a proof-of-principle, we show that for a two-parameter model that depends only on the thermally-averaged annihilation cross section ((o -v)) and the dark-matter mass (Mx), QCD uncertainties can affect the best-fit mass by up to ti 14% (with large uncertainties for large DM masses), depending on the choice of Mx and the annihilation channel (bb over bar or W+W-), and (o -v) by up to ti 10%. For comparison, changes to the underlying diffusion parameters are found to be within 1%-5%, and the results are also quite resilient to the choice of cosmic-ray propagation model. These findings indicate that QCD uncertainties need to be included in future DM analyses. To facilitate full-fledged analyses, we provide the spectra in tabulated form including QCD uncertainties and code snippets to perform mass interpolations and quick DM fits. The code can be found in this GitHub [1] repository.

Abstract: We propose a minimal gauged U(1)(B-L) extension of the minimal supersymmetric Standard Model (MSSM) which resolves the cosmological moduli problem via thermal inflation, and realizes late-time Affleck-Dine leptogensis so as to generate the right amount of baryon asymmetry at the end of thermal inflation. The present relic density of dark matter can be explained by sneutrinos, MSSM neutralinos, axinos, or axions. Cosmic strings from U(1)(B-L) breaking are very thick, and so the expected stochastic gravitational wave background from cosmic string loops has a spectrum different from the one in the conventional Abelian-Higgs model, as would be distinguishable at least at LISA and DECIGO. The characteristic spectrum is due to a flat potential, and may be regarded as a hint of supersymmetry. Combined with the resolution of moduli problem, the expected signal of gravitational waves constrains the U(1)(B-L) breaking scale to be O(10(12-13)) GeV. Interestingly, our model provides a natural possibility for explaining the observed ultra-high-energy cosmic rays thanks to the fact that the core width of strings in our scenario is very large, allowing a large enhancement of particle emissions from the cusps of string loops. Condensation of LHu flat-direction inside of string cores arises inevitably and can also be the main source of the ultra-high-energy cosmic rays accompanied by ultra-high-energy lightest supersymmetric particles.