Artificial homeostatic temperature regulation via bio-inspired feedback mechanisms.

Homeostasis comprises one of the main features of living organisms that enables their robust functioning by adapting to environmental changes. In particular, thermoregulation, as an instance of homeostatic behavior, allows mammals to maintain stable internal temperature with tightly controlled self-regulation independent of external temperatures. This is made by a proper reaction of the thermoeffectors (like skin blood vessels, brown adipose tissue (BAT), etc.) on a wide range of temperature perturbations that reflect themselves in the thermosensitive neurons' activity. This activity is being delivered to the respective actuation points and translated into thermoeffectors' actions, which bring the temperature of the organism to the desired level, called a set-point. However, it is still an open question whether these mechanisms can be implemented in an analog electronic device: both on a system theoretical and a hardware level. In this paper, we transfer this control loop into a real electric circuit by designing an analog electronic device for temperature regulation that works following bio-inspired principles. In particular, we construct a simplified single-effector regulation system and show how spiking trains of thermosensitive artificial neurons can be processed to realize an efficient feedback mechanism for the stabilization of the a priori unknown but system-inherent set-point. We also demonstrate that particular values of the set-point and its stability properties result from the interplay between the feedback control gain and activity patterns of thermosensitive artificial neurons, for which, on the one hand, the neuronal interconnections are generally not necessary. On the other hand, we show that such connections can be beneficial for the set-point regulation and hypothesize that the synaptic plasticity in real thermosensitive neuronal ensembles can play a role of an additional control layer empowering the robustness of thermoregulation. The electronic realization of temperature regulation proposed in this paper might be of interest for neuromorphic circuits which are bioinspired by taking the basal principle of homeostasis on board. In this way, a fundamental building block of life would be transferred to electronics and become a milestone for the future of neuromorphic engineering.

Structural plasticity driven by task performance leads to criticality signatures in neuromorphic oscillator networks.

Oscillator networks rapidly become one of the promising vehicles for energy-efficient computing due to their intrinsic parallelism of execution. The criticality property of the oscillator-based networks is regarded to be essential for performing complex tasks. There are numerous bio-inspired synaptic and structural plasticity mechanisms available, especially for spiking neural networks, which can drive the network towards the criticality. However, there is no solid connection between these self-adaption mechanisms and the task performance, and it is not clear how and why particular self-adaptation mechanisms contribute to the solution of the task, although their relation to criticality is understood. Here we propose an evolutionary approach for the structural plasticity that relies solely on the task performance and does not contain any task-independent adaptation mechanisms, which usually contribute towards the criticality of the network. As a driver for the structural plasticity, we use a direct binary search guided by the performance of the classification task that can be interpreted as an interaction of the network with the environment. Remarkably, such interaction with the environment brings the network to criticality, although this property was not a part of the objectives of the employed structural plasticity mechanism. This observation confirms a duality of criticality and task performance, and legitimizes internal activity-dependent plasticity mechanisms from the viewpoint of evolution as mechanisms contributing to the task performance, but following the dual route. Finally, we analyze the trained network against task-independent information-theoretic measures and identify the interconnection graph's entropy to be an essential ingredient for the classification task performance and network's criticality.

On the interpretation of Dirac δ pulses in differential equations for phase oscillators.

In this note, we discuss the usage of the Dirac δ function in models of phase oscillators with pulsatile inputs. Many authors use a product of the delta function and the phase response curve in the right-hand side of an ordinary differential equation to describe the discontinuous phase dynamics in such systems. We point out that this notation has to be treated with care as it is ambiguous. We argue that the presumably most canonical interpretation does not lead to the intended behavior in many cases.