A fast algorithm to the conjugacy problem on generic braids

Abstract

Random braids that are formed by multiplying randomly chosen permutation braids are studied by analyzing their behavior under Garside's weighted decomposition and cycling. Using this analysis, we propose a polynomial-time algorithm to the conjugacy problem that is successful for random braids in overwhelming probability. As either the braid index or the number of permutation-braid factors increases, the success probability converges to 1 and so, contrary to the common belief, the distribution of hard instances for the conjugacy problem is getting sparser. We also prove a conjecture by Birman and González-Meneses that any pseudo-Anosov braid can be made to have a special weighted decomposition after taking power and cycling. Moreover we give polynomial upper bounds for the power and the number of iterated cyclings required.