Basin Geometry and Reliable Recall of Dynamical Memories in Reservoir Computing

2026-09-01

AI Summary

研究解决了在储层计算中可靠召回动态记忆的问题，通过揭示记忆盆地的‘章鱼状’结构，并利用线索驱动的广义同步机制来克服不可预测性。

Abstract

Reliable attractor recall conventionally requires broad basins of attraction. However, in reservoir-computing based associative memory, temporal cues reliably recover dynamical memories despite basins dominated by unpredictable, riddled-like regions. We reveal that memory basins exhibit an ``octopus-like'' structure: a robust ``head'' near the attractor and thin, intertwined ``tentacles'' spanning state space. Initial states in tentacular regions yield near-zero uncertainty exponents, making the recalled memory effectively unpredictable at finite precision. Yet, cue-driven generalized synchronization bypasses this unpredictability, driving the system into the robust basin head. This mechanism yields a quantitative relation linking minimum cue duration, synchronization rate, and basin-head radius. Trained recurrent neural networks exhibit similar geometry, suggesting this phenomenon extends beyond reservoir computing.

Problem

• basin of attraction
• reservoir computing
• associative memory
• dynamical memories
• generalized synchronization

Innovation

• octopus-like structure
• generalized synchronization
• attractor recall
• basin geometry
• reservoir computing