Phasor Agents: Oscillatory Graphs with Three-Factor Plasticity and Sleep-Staged Learning

TL;DR

Phasor Agents utilize oscillatory graph dynamics with three-factor plasticity and sleep-staged learning to achieve stable, biologically inspired learning and memory consolidation, demonstrating significant improvements in stability and performance in complex tasks.

Contribution

This work introduces a novel oscillatory graph system with three-factor plasticity and sleep-inspired mechanisms for stable learning without backpropagation.

Findings

Eligibility traces preserve credit under delayed modulation

Sleep-stage dynamics improve stability and learning capacity

Replay enhances maze success rate by +45.5 percentage points

Abstract

Phasor Agents are dynamical systems whose internal state is a Phasor Graph: a weighted graph of coupled Stuart-Landau oscillators. A Stuart-Landau oscillator is a minimal stable "rhythm generator" (the normal form near a Hopf bifurcation); each oscillator is treated as an abstract computational unit (inspired by, but not claiming to model, biological oscillatory populations). In this interpretation, oscillator phase tracks relative timing (coherence), while amplitude tracks local gain or activity. Relative phase structure serves as a representational medium; coupling weights are learned via three-factor local plasticity - eligibility traces gated by sparse global modulators and oscillation-timed write windows - without backpropagation. A central challenge in oscillatory substrates is stability: online weight updates can drive the network into unwanted regimes (e.g., global synchrony), collapsing representational diversity. We therefore separate wake tagging from offline consolidation, inspired by synaptic tagging-and-capture and sleep-stage dynamics: deep-sleep-like gated capture commits tagged changes safely, while REM-like replay reconstructs and perturbs experience for planning. A staged experiment suite validates each mechanism with ablations and falsifiers: eligibility traces preserve credit under delayed modulation; compression-progress signals pass timestamp-shuffle controls; phase-coherent retrieval reaches 4x diffusive baselines under noise; wake/sleep separation expands stable learning by 67 percent under matched weight-norm budgets; REM replay improves maze success rate by +45.5 percentage points; and a Tolman-style latent-learning signature - immediate competence and detour advantage after unrewarded exploration, consistent with an internal model - emerges from replay (Tolman, 1948). The codebase and all artifacts are open-source.