A challenge for researchers developing applications of neuromorphic [or human brain-inspired] hardware is that a formal hierarchy such as Turing completeness does not currently exist. Instead, each new chip architecture requires a custom software toolchain — a set of programming tools — that defines algorithms and executes them by mapping them onto the unique hardware. This makes it difficult to compare the performance of different neuromorphic systems executing the same algorithm, and requires researchers to understand all aspects of the algorithm and hardware to obtain the potentially brain-like performance.[Youhui] Zhang et al. now present a breakthrough solution to this problem by proposing a concept that they call neuromorphic completeness — which, in a nod to Turing completeness, aims to decouple algorithm and hardware development. In a relaxation of the hierarchy for conventional computers, the authors propose that a brain-inspired system is neuromorphic complete if it can execute a given set of fundamental operations with a prescribed level of accuracy (Fig. 1). This is a deviation from Turing completeness, in which a system can be defined as complete only if it provides an exact and equivalent result for a given set of fundamental operations.
It remains to be seen whether actual brains — biological ‘hardware’ — are themselves neuromorphic complete, but the authors’ approach nevertheless brings us closer to the great gains that could be made using brain-inspired hardware.