DPR框架
Year: 2026
Type: preprint
Abstract: This paper proposes a unified foundational structure—the Difference-Protocol-Response framework (DPR framework)—for describing the emergence, evolution, and nesting of structure in any world that contains internal differences. The framework rests on four axioms: the Axiom of Distinguishability establishes the most primitive form of difference as a distinction (A | ¬A); the Axiom of Protocol Co-genesis asserts that distinction and protocol are two faces of a single event, requiring no first cause; the Axiom of Iteration and Interaction provides three mechanisms—nesting, temporal ordering, and boundary interaction—that transform an isolated distinction into a self-organizing DPR network; the Axiom of Finite Describability restricts the theory to realizable structures. Within this framework, "computation" is dissolved—a Turing machine is nothing more than the mechanical flow of difference-protocol-response, and "computation" is the semantic label an observer attaches to compressible DPR chains. Processing falls into two modes: F-type (Full Processing, executing low-level chains without compression) and C-type (Compressed Processing, encoding low-level outputs as high-level singular differences). Intelligence is redefined as the capacity to construct multi-level C-type compression protocols in a computationally irreducible world—and intelligence cannot be neutral, because every act of compression is a value-laden choice about which differences matter. The framework spans abstraction and reality: at the abstract end, it is deeply homologous with category theory (difference ≈ object, protocol ≈ morphism); at the concrete end, it accommodates the layered distinctions among physical necessity, evolutionary lock-in, and protocol contingency. The paper further applies the DPR framework to open problems including computational irreducibility, P versus NP, the Church-Turing thesis, and the necessity versus contingency of mathematical foundations, and proposes scaling laws as a candidate direction for moving from qualitative description to quantitative prediction. 本文提出一个统一的底层结构——差异-协议-响应框架(DPR框架),用以描述任何包含内部差异的世界中结构的产生、演化与嵌套。框架以四个公理为基础:可区分性公理确立差异的最原初形式为一个区分(A | ¬A);协议共生公理断言区分与协议是一体两面,无需第一因;迭代与交互公理提供嵌套、时序和边界交互三个机制,使孤立区分扩展为自组织的DPR网络;有限描述公理将理论限定于可实现的结构。在这一框架下,"计算"被消解——图灵机不过是差异-协议-响应的机械流转,"计算"是可压缩DPR链被观察者贴上的语义标签。处理模式分为F型(完整执行,不做压缩)和C型(压缩低层输出为高层单一差异),智能被重新定义为在计算不可归约的世界中构建多层C型压缩协议的能力——且智能不可能中立,因为每一次压缩都是有立场的价值判断。框架贯通抽象与现实:在抽象端与范畴论高度同源(差异≈对象,协议≈态射),在现实端兼容物理约束、演化锁定和协议偶然的层级区分。本文进一步将DPR框架应用于计算不可归约性、P vs NP问题、丘奇-图灵论题、数学基础的必然性与偶然性等开放问题,并提出标度律作为从定性走向定量的候选方向。
Author: 赵, 磊
Language: English
Subfield: Computational Theory and Mathematics
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7171537896
Cites:
Cited by:
Related to: