Reply to: Limitations in odour recognition and generalisation in a neuromorphic olfactory circuit

TL;DR

This paper defends previous findings on neuromorphic olfactory circuits by addressing critiques about dataset limitations, generalization ability, and distance measures, clarifying the robustness of their original conclusions.

Contribution

The authors clarify how their original study effectively addressed dataset limitations and demonstrate the robustness of their neuromorphic olfactory model despite critiques.

Findings

Dataset limitations do not undermine original conclusions

The neuromorphic EPL network generalizes well over repeated odor presentations

Their performance claims are not matched by the alternative distance measure

Abstract

Dennler et al. submit that they have discovered limitations affecting some of the conclusions drawn in our 2020 paper, Rapid online learning and robust recall in a neuromorphic olfactory circuit. Specifically, they assert (1) that the public dataset we used suffers from sensor drift and a nonrandomized measurement protocol, (2) that our neuromorphic EPL network is limited in its ability to generalize over repeated presentations of an odorant, and (3) that our EPL network results can be performance matched by using a more computationally efficient distance measure. Though they are correct in their description of the limitations of that public dataset, they do not acknowledge in their first two assertions how our utilization of those data sidestepped these limitations. Their third claim arises from flaws in the method used to generate their distance measure. We respond below to each of these three claims in turn.