arXiv cs.AI· Daniel Levine, David van Dijk, Kenneth Kay·· 1 天前AI 评分31
AutoHebb：可零样本推理与持续学习的自修改神经网络
AutoHebb: Self-Modifying Neural Networks that Zero-Shot Infer and Continually Learn
AI 导读
研究者提出 AutoHebb，一类在线学习的前馈神经网络，通过外部反馈与内部生成活动产生全局学习信号（神经调质），门控三因子 Hebbian 局部学习规则来更新可塑性权重。借助元学习搜索固定权重后，AutoHebb 在零样本推理任务上成功，并涌现出未经训练的零样本推理与持续学习能力，表现优于现代学习模型。消融对比显示元学习、Hebbian 可塑性和学习自主控制三者缺一不可。
正文 · 原文
Abstract:Humans and animals have a unique capacity to generalize and learn in novel circumstances, including "0-shot" inference (ZI) and continual learning (CL) -- two powerful cognitive abilities. Despite this importance, efforts to model ZI and CL are seriously limited in several ways. First, existing models typically lack components critical to ZI and CL in the brain, namely, strong inductive biases (for ZI) and autonomous control of learning (for CL). Second, models can be less tractable to analysis if model components are overly complex (lacking simplicity), or too task-restricted if model components are overly bespoke (lacking expressivity). Here we introduce autonomous Hebbian networks ("AutoHebb"), a class of online learning feedforward neural networks that realize strong inductive biases and autonomously controlled learning, and are both simple and expressive. Unlike standard models that learn via externally controlled optimization (e.g. gradient descent), AutoHebb networks combine external feedback with internally generated activity to produce a global learning signal ('neuromodulator') fed back to the network. The neuromodulator in turn gates a biologically grounded local learning rule (3-factor Hebbian rule) that updates networks' plastic weights. Using meta-learning to search for fixed (non-plastic) weights, we find that AutoHebb networks (1) succeed on ZI tasks, (2) show emergent (untrained) ZI/CL abilities, (3) admit reverse-engineering into explicit algorithms, and (4) outperform modern learning models on ZI/CL tasks. Lastly, performance comparisons to related models implied an essential role for three model components: meta-learning, Hebbian plasticity, and autonomous control of learning. These findings indicate that AutoHebb may be useful in hypothesis generation, analysis, and discovery of biologically grounded forms of learning and generalization.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE) |
| Cite as: | arXiv:2610.40757 [cs.LG] |
| (or arXiv:2610.40757v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.40757 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Daniel Levine [view email]
[v1]
Thu, 29 Oct 2026 16:57:55 UTC (2,949 KB)
来源：arXiv cs.AI · arxiv.org