Flavours of Physics was a machine learning competition, hosted by Kaggle and partly sponsed by CERN. The goal of the competition was to gain sensitivity in the search for τ → 3μ decays. My entry ranked 80 out of 673 teams on the private leaderboard. I will explain my methodology and discuss some random thoughts about the competition.
##τ → 3μ decays? It is a well known fact that lepton flavor(LF) numbers are strictly conserved in the Standard Model (SM). However, observations from certain neutrino experiments strongly suggests that there is some lepton flavor mixing as well. What this means is that charged lepton flavor violating (cLFV) processes may also occur at some level. Many current experiments are paying keen attention in the search for LFV processes in the charged lepton sector such as tau decays, Z → μτ decays, etc. LFV observed in experiments would be evidence of new physics beyond the SM. And that in essence was the basis of the basis of this competition.
##Opening remarks This was my first ever machine learning competition that I've participated in. Initially, I found it quite daunting and noticed my knowledge of machine learning, in general, was quite inadequate. However, over the two month span of the competition, I can gladly say that I've learnt an enormous amount.