受生物启发，用进化算法优化感知架构，提升长期避损学习效率。
Evolving Afferent Architectures: Biologically-inspired Models for Damage-Avoidance Learning
- 通过双层架构：进化优化选感知结构，强化学习训练避损策略
- 在长期生物力学数字孪生中实现23%高风险行为减少，更具年龄适应性
- 适合研究长期动态系统、生物启发学习与可解释决策的学者
我们提出一种名为传入学习（Afferent Learning）的框架，生成计算传入信号（CATs）作为内部风险信号，用于损伤规避学习。该框架受生物系统启发，采用双层结构：外层通过进化优化发现能有效支持策略学习的传入感知架构；内层利用这些信号进行强化学习，训练损伤规避策略。这将传入感知形式化为促进高效学习的归纳偏置——架构选择基于其对学习的促进能力，而非直接最小化损伤。在具有挑战性的长期生物力学数字孪生场景中（跨越数十年生命周期），我们发现基于CAT的进化架构显著提升了学习效率与年龄鲁棒性，使策略实现年龄依赖的行为自适应，高风险行为降低23%。消融实验验证了CAT信号、进化机制和预测差异的关键作用。代码与数据已开源，支持复现。
原文摘要 · Abstract (English)
We introduce Afferent Learning, a framework that produces Computational Afferent Traces (CATs) as adaptive, internal risk signals for damage-avoidance learning. Inspired by biological systems, the framework uses a two-level architecture: evolutionary optimization (outer loop) discovers afferent sensing architectures that enable effective policy learning, while reinforcement learning (inner loop) trains damage-avoidance policies using these signals. This formalizes afferent sensing as providing an inductive bias for efficient learning: architectures are selected based on their ability to enable effective learning (rather than directly minimizing damage). We provide theoretical convergence guarantees under smoothness and bounded-noise assumptions. We illustrate the general approach in the challenging context of biomechanical digital twins operating over long time horizons (multiple decades of the life-course). Here, we find that CAT-based evolved architectures achieve significantly higher efficiency and better age-robustness than hand-designed baselines, enabling policies that exhibit age-dependent behavioral adaptation (23% reduction in high-risk actions). Ablation studies validate CAT signals, evolution, and predictive discrepancy as essential. We release code and data for reproducibility.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。