maya-metrics: A Stateless Python Library for Affective Neuromorphic Evaluation in Continual Learning SNNs

Key Points

• To develop and validate maya-metrics, an open-source library for assessing affective dynamics in neuromorphic networks during continual learning.
• Developed six stateless modules: AffectiveMetrics, CrossDimensional, ComplexityMetrics, MaturationIndex, CrossSubstrate, and CLCorrector.
• Validated against 11 published studies across two substrates: spiking neural networks and large language models.
• Conducted a unit test with 16/16 passing successfully.
• Confirmed two constants: Bhaya Quiescence Law and Buddhi S-curve determinism.
• Demonstrated the effectiveness of the library for evaluating class-incremental learning in SNNs.

Abstract