AI with Symbolic Empathy: Shannon-Neumann Insight Guided Logic

2026/01/03 by Edouard Siregar · 1 voice
Computer Science · Psychology · Neuroscience · #Explainable Artificial Intelligence (XAI) #Child and Animal Learning Development #Embodied and Extended Cognition

paper · pdf · doi:10.1007/s12559-025-10536-9

openalex publication_date 2026/01/03 · openalex created_date 2026/01/04 · openalex updated_date 2026/07/29

Abstract

We present an Artificial Intelligence with Symbolic Empathy, where an agent \mathcal A cooperatively aligns a person’s cognitive state X_P(t) with their ideal trajectory X_LP(t) . Alignment is guided by a context-sensitive, non-monotonic logic \mathcal L_\mathcal A implemented as a five-stage Hierarchical FSM: \textsfIdentify (fact structuring), \textsfClassify (pattern extraction), \textsfExplore (intervention generation), \textsfPlan (ontology-based misalignment diagnosis), and \textsfReason (abductive hypothesis evaluation). Reasoning is driven by the Shannon-von Neumann insight gain I_SN(γ _i) = [ \mathcal H(B_P(t)) - \mathcal H(B_P⁽ⁱ⁾(t+1)) ] ⋅ U(γ _i), combining entropy reduction with goal-relevant utility. This metric enables self-supervised abductive learning, refining \mathcal A ’s theory-of-mind and guiding psychologically meaningful, uncertainty-reducing interventions. The architecture ensures interpretability (I-AI), explainability (X-AI), and trustworthiness (T-AI) via explicit fact-rule tracing. Cognitive-cooperative case studies show that integrating symbolic causal models with utility-guided abductive-deductive reasoning yields transparent, goal-aligned human-AI interaction.

Citations

• A note on the physical interpretation of neural PDE's
• Algebraic Dynamical Systems in Machine Learning
• GPT-4 Technical Report
• Flamingo: a Visual Language Model for Few-Shot Learning
• Scaling Laws for Neural Language Models
• Attention Is All You Need
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Scaling Memory-Augmented Neural Networks with Sparse Reads and Writes
• The Mythos of Model Interpretability
• Neural Machine Translation of Rare Words with Subword Units
• Efficient Estimation of Word Representations in Vector Space

Discussions

• Research : doi.org/10.1007/s125... @springernature.com #aimentalhealthethicsresearch #aiwellbeingtechnologyframework #compassionateartificialintelligence #humancentredaidesign #safeinterpretablea [bsky, 0 points, 0 comments]