深度学习2.0：真正重要的人工神经元——拒绝相关性，拥抱正交性
Deep Learning 2.0: Artificial Neurons That Matter -- Reject Correlation, Embrace Orthogonality
AI总结：
该研究提出基于yat乘积的神经物质网络（NMN），无需激活函数即可实现非线性模式识别，简化架构并提升决策透明度，在多数据集上持续优于传统MLP，挑战激活函数必要性假设并建立新设计范式。
摘要：
我们提出一种基于yat乘积的神经网络，即神经物质网络（NMN），这是深度学习领域的一项突破，无需激活函数即可实现非线性模式识别。我们的核心创新依赖于yat乘积，它通过将输入投影到伪度量空间自然引入非线性，消除了对传统激活函数的需求，同时仅保留softmax层以输出最终类概率分布。该方法简化了网络架构，并为网络决策过程提供了前所未有的透明度。我们在不同数据集上的综合实证评估表明，NMN持续优于传统MLP。研究结果挑战了“独立激活函数对有效深度学习模型必不可少”的假设。这项工作的意义超出了直接的架构优势：通过消除中间激活函数同时保留非线性能力，yat-MLP建立了一种兼具简单性和有效性的神经网络设计新范式。最重要的是，我们的方法为传统上不透明的神经网络“黑箱”特性提供了前所未有的见解，有助于更清晰地理解这些模型如何处理和分类信息。
英文摘要：
We introduce a yat-product-powered neural network, the Neural Matter Network (NMN), a breakthrough in deep learning that achieves non-linear pattern recognition without activation functions. Our key innovation relies on the yat-product and yat-product, which naturally induces non-linearity by projecting inputs into a pseudo-metric space, eliminating the need for traditional activation functions while maintaining only a softmax layer for final class probability distribution. This approach simplifies network architecture and provides unprecedented transparency into the network's decision-making process. Our comprehensive empirical evaluation across different datasets demonstrates that NMN consistently outperforms traditional MLPs. The results challenge the assumption that separate activation functions are necessary for effective deep-learning models. The implications of this work extend beyond immediate architectural benefits, by eliminating intermediate activation functions while preserving non-linear capabilities, yat-MLP establishes a new paradigm for neural network design that combines simplicity with effectiveness. Most importantly, our approach provides unprecedented insights into the traditionally opaque "black-box" nature of neural networks, offering a clearer understanding of how these models process and classify information.
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。
