Create and explore imaginative 3D worlds with our new product Dreamfield (opens in a new tab)

Local Plasticity Rules for Desired Network-Level Behavior in Spiking Neural Networks

Identify the local synaptic and structural plasticity rules that yield a specified network-level behavior in spiking neural networks implemented on neuromorphic hardware, thereby linking local learning mechanisms to emergent system-level functionality.

Background

The paper discusses the need for on-device and online learning in neuromorphic systems, emphasizing local plasticity as a promising approach. However, due to emergent dynamics, deriving system-level behaviors from local rules remains challenging.

The authors explicitly state that it is not clear which local rules produce particular global behaviors, noting that evolutionary search and meta-learning have been used to rediscover useful rules but without establishing a general mapping from rules to behaviors.

References

It is not clear what local rules will yield a particular network-level behavior, but evolutionary search and meta-learning have been used to (re-)discover desirable plasticity rules.

— Neuromorphic Programming: Emerging Directions for Brain-Inspired Hardware  (2410.22352 - Abreu et al., 2024) in Section 5 (Neuromorphic Programming), Online learning paragraph

One remaining hypothesis is a mismatch between the states represented during reward-producing bursts and those encountered at later selections; this explanation remains untested.

— The Fly That Stopped: Mushroom-Body-Inspired Habituation as a Reward-Free Scheduling Prior for Autonomous Penetration Testing  (2609.29126 - Moutesidis, 24 Sep 2026) in Section 6.2, “Five local-plasticity approaches under a fixed readout”

In particular, it remains to be clarified which functional advantages arise from sparse and selectively structured connectivity rather than all-to-all communication (for example through modularization, reduced population-wide correlations), plasticity operating on slower timescales, and whether the apparent stochasticity in rewiring may itself have functional advantages (for example by enabling the sampling of alternative connectivity configurations).

— Rewiring of adult neural networks - learning from structural plasticity across multiple scales  (2610.18024 - Fauth et al., 13 Oct 2026) in Section 1, subsection “Open problems,” paragraph “What are the effects on the network level?”

A further open question is how structural plasticity at the single neuron scale influences activity and function on the scale of large networks or even whole brains.

— Rewiring of adult neural networks - learning from structural plasticity across multiple scales  (2610.18024 - Fauth et al., 13 Oct 2026) in Section 2, subsection “Open problems,” paragraph “The influence of rewiring on network and whole brain activity”