Neuromodulated Equilibrium Learning
- 1Lakehead University
Abstract
Equilibrium Propagation (EP) is a biologically motivated alternative to backpropagation but requires free and nudged equilibrium phases, raising questions about the biological plausibility of phase-separated learning. We introduce Neuromodulated Equilibrium Learning (NEL), a single-equilibrium reward-based framework inspired by Attention-Gated Brain Propagation and neuromodulatory signaling. After reaching a free equilibrium, NEL propagates a reward-dependent modulatory signal derived from the selected action through reciprocal feedback connections to drive local synaptic updates, eliminating both the reward-nudged equilibrium and contrastive update. We evaluate NEL on MNIST, Fashion-MNIST, and CIFAR-10 using multilayer perceptrons and convolutional networks, with symmetric three-phase REP and standard EP as multi-phase baselines. NEL approaches three-phase REP in shallow architectures, with gaps of 0.79 and 1.31 percentage points on MNIST and one-hidden-layer Fashion-MNIST, while larger gaps of 3.13 and 6.21 percentage points emerge in the deeper Fashion-MNIST and CIFAR-10 settings. A direct cost feedback control using the full supervised output error also remains below three-phase EP, indicating that sparse reward information alone does not explain the performance gap between single-equilibrium and multi-phase learning. These results show that effective reward-driven learning can arise from a single free equilibrium while suggesting that computations introduced by additional nudged relaxations become increasingly important in more challenging architectures.
Related
- dITP shifts the assembly equilibrium of Kongming to activate NAD+ hydrolysis
- Thermodynamics of consciousness: Non-equilibrium brain dynamics track conscious states
- Prospective Coding and Path Integration Emerge as Equilibrium Solutions of Self-Organizing Neural Networks with Firing-Rate Adaptation
- Immersed Boundary Electrodiffusion in the Wasserstein Gradient Flow Formalism: Global Well-Posedness, Non-Equilibrium Thermodynamics, and Singular Asymptotics
- ROS Equilibrium in the Digestive Fluid of the Nepenthes × ventrata Trap at Various Developmental Stages.
BibTeX
@misc{kubo2026neuromodulated,
title = {{Neuromodulated Equilibrium Learning}},
author = {Y. Kubo},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.28.747901},
url = {https://www.biorxiv.org/content/10.64898/2026.08.28.747901v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Neuromodulated Equilibrium Learning
AU - Y. Kubo
PY - 2026
DA - 2026/08/28
PB - bioRxiv
DO - 10.64898/2026.08.28.747901
UR - https://www.biorxiv.org/content/10.64898/2026.08.28.747901v1
AB - Equilibrium Propagation (EP) is a biologically motivated alternative to backpropagation but requires free and nudged equilibrium phases, raising questions about the biological plausibility of phase-separated learning. We introduce Neuromodulated Equilibrium Learning (NEL), a single-equilibrium reward-based framework inspired by Attention-Gated Brain Propagation and neuromodulatory signaling. After reaching a free equilibrium, NEL propagates a reward-dependent modulatory signal derived from the selected action through reciprocal feedback connections to drive local synaptic updates, eliminating both the reward-nudged equilibrium and contrastive update. We evaluate NEL on MNIST, Fashion-MNIST, and CIFAR-10 using multilayer perceptrons and convolutional networks, with symmetric three-phase REP and standard EP as multi-phase baselines. NEL approaches three-phase REP in shallow architectures, with gaps of 0.79 and 1.31 percentage points on MNIST and one-hidden-layer Fashion-MNIST, while larger gaps of 3.13 and 6.21 percentage points emerge in the deeper Fashion-MNIST and CIFAR-10 settings. A direct cost feedback control using the full supervised output error also remains below three-phase EP, indicating that sparse reward information alone does not explain the performance gap between single-equilibrium and multi-phase learning. These results show that effective reward-driven learning can arise from a single free equilibrium while suggesting that computations introduced by additional nudged relaxations become increasingly important in more challenging architectures.
ER -