研究人员引入了 EML（Exp-Minus-Log）树，这是一种新颖的组合构建模块，作为 NAND 门的连续模拟。这些 EML 树在 $W^{k, \infty}$ 空间内展现出函数的通用逼近性质，与经典的神经网络逼近论证有相似之处。该研究还提出了一种实用的 EML 树学习算法，展示了它们在优化问题中的潜力，并确立了它们作为函数逼近的理论上可靠的框架。 AI
排序理由 该集群包含一篇详细介绍函数逼近新理论框架的学术论文。
在 arXiv cs.NE (Neural & Evolutionary) 阅读 →
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
The recently introduced EML (Exp-Minus-Log) function acts as continuous analogue of NAND gates, providing a compositional building block capable of representing elementary functions. In this work, we study the expressive power of tree-structured compositions of EML functions. We …
The recently introduced EML (Exp-Minus-Log) function acts as continuous analogue of NAND gates, providing a compositional building block capable of representing elementary functions. In this work, we study the expressive power of tree-structured compositions of EML functions. We …