arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.01393cond-mat.dis-nn
收藏

从噪声示例中推断概念的Hopfield类神经网络

Inferring Concepts from Noisy Examples in Hopfield-like Neural Networks

Università Bocconi(博科尼大学);Sorbonne Université, CNRS, Laboratoire de Physique Théorique de la Matière Condensée(索邦大学,法国国家科学研究中心,凝聚态物质理论物理实验室);Ca’ Foscari Università Venezia(威尼斯卡福斯卡里大学);ZHAW Switzerland(苏黎世应用科学与艺术大学);Università di Roma La Sapienza(罗马第一大学)

Marco Benedetti, Victor Dotsenko, Giulia Fischetti, Enzo Marinari, Gleb Oshanin

AI总结:

该研究通过Hopfield类神经网络的伪逆学习规则变体,揭示了在噪声示例下推断概念的机制,并展示了更强的副本对称性破缺效应。

摘要:

我们研究了一种Hopfield类神经网络的伪逆学习规则变体,该规则使网络能够在有限数量的示例基础上推断典型概念。该模型的均场副本理论揭示了这种泛化能力如何通过多种具有不同热力学性质的状态共同存在来实现,这些状态与标准Hopfield状态共存。它们通过平滑的过渡或不连续的跳跃出现和消失,有趣的是,它们表现出比标准Hopfield模型强得多的副本对称性破缺(RSB)效应,这通过我们的1RSB分析得以捕捉。我们的结果与数值模拟高度一致,为吸引子神经网络中记忆存储与泛化之间的相互作用提供了更深入的理解。

英文摘要:

We study a variant of the pseudo-inverse learning rule for Hopfield-like Neural Networks, which allows the network to infer archetypal concepts on the basis of a limited number of examples. The key novelty is that the new learning rule compensates for correlations between examples at the population level rather than at the level of the particular training set, reducing overfitting of the dataset and leading to generalization. The mean-field replica theory for this model reveals how this generalization ability is mediated by a multitude of thermodynamic states, with diverse thermodynamic properties, coexisting with the standard Hopfield ones. They appear and vanish through smooth transitions or discontinuous jumps and, interestingly, show much stronger Replica Symmetry Breaking (RSB) effects than the standard Hopfield model, as captured by our 1RSB analysis. Our results, in excellent agreement with numerical simulations, provide deeper insight into the interplay between memory storage and generalization in attractor neural networks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑