arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.18090cs.PL
收藏

使用箭头反向处理器的反向传播编程

Programming Backpropagation with Reverse Handlers for Arrows

Fukui Prefectural University(福井县立大学);Kyoto University(京都大学);Kyoto Sangyo University(京都产业大学)

Takahiro Sanada, Keisuke Hoshino, Kenshin Hirai, Shin-ya Katsumata

AI总结:

提出一种基于代数效应和箭头处理器的编程语言,通过反向处理器实现神经网络的反向传播,提供抽象网络描述和灵活实现,并建立类型系统、操作语义和范畴语义。

摘要:

我们引入了一种新的编程语言及其范畴语义,以便在代数效应和箭头处理器的框架内设计和实现神经网络。我们的语言使我们能够以与代数效应相同的方式符号化地构建神经网络,并通过处理器为它们分配实现——例如反向传播计算。这种语言设计的优势在于网络描述变得抽象且高层,而实现可以灵活地分配给网络。我们通过开发类型系统、操作语义、范畴语义以及可靠性和充分性定理,为我们的语言建立了严格的基础。技术核心是引入了反向处理器,这是一种用于实现反向传播的箭头的新型处理器机制,以及反向微分限制范畴(RDRCs)上的强promonad的新代数,其字符串图为神经网络提供了形式化的图形语法和语义。

英文摘要:

We introduce a new programming language and its categorical semantics in order to design and implement neural networks within the framework of algebraic effects and handlers for arrows. Our language enables us to construct neural networks symbolically, in the same manner as algebraic effects, and to assign implementations -- such as backpropagation computations -- to them via handlers. The advantage of this language design is that network descriptions become abstract and high-level, while implementations can be flexibly assigned to networks. We establish a rigorous foundation for our language by developing a type system, an operational semantics, a categorical semantics, and soundness and adequacy theorems. The technical core is the introduction of reverse handlers, a novel handler mechanism for arrows for implementing backpropagation, together with new algebras of strong promonads on reverse differential restriction categories (RDRCs), whose string diagrams provide a formal graphical syntax and semantics for 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。

↑