2. Topographic representation for quantum machine learning
Bruce James MacLennan · 2020
One of the most common information representations in the brain is the topographic or computational map, in which neurons are arranged systematically according to the values they represent. By representing quantitative relationships spatially, computational maps enable the brain to compute complex, nonlinear functions to the accuracy required. This chapter proposes two approaches to quantum computation for machine learning by means of topographic representation. It shows how to construct unitary operators, implementable on quantum computers, that implement arbitrary (including nonlinear) functions via computational maps.