paper-with-me

Papers

Self-Modifying Code in Open-Ended Evolutionary Systems

2022-01-18 · Patrik Christen

Having a model and being able to implement open-ended evolutionary systems is important for advancing our understanding of open-endedness. Complex systems science and newest generation high-level programming languages provide intriguing possibilities to do so. First, some recent advances in modelling and implementing open-ended evolutionary systems are reviewed. Then, the so-called allagmatic method is introduced that describes, models, implements, and allows interpretation of complex systems. After highlighting some current modelling and implementation challenges, model building blocks of open-ended evolutionary systems are identified, a system metamodel of open-ended evolution is formalised in the allagmatic method, an implementation self-modifying code prototype with a high-level programming language is provided, and guidance from the allagmatic method to create code blocks is described. The proposed prototype allows modifying code at runtime in a controlled way within a system metamodel. Since the allagmatic method has been built based on metaphysical concepts borrowed from Gilbert Simondon and Alfred N. Whitehead, the proposed prototype provides a promising starting point to interpret novelty generated at runtime with the help of a metaphysical framework.

📄 PDF Abstract BibTeX arXiv:2201.06858

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

DéjàQ: Open-Ended Evolution of Diverse, Learnable and Verifiable Problems

2026-01-05 · Willem Röpke, Samuel Coward, Andrei Lupu, Thomas Foster 외 arxiv

Recent advances in reasoning models have yielded impressive results in mathematics and coding. However, most approaches rely on static datasets, which have been suggested to encourage memorisation and limit generalisatio…

Mathematical Reasoning

SePO: Self-Evolving Prompt Agent for System Prompt Optimization

2026-06-03 · Wangcheng Tao, Han Wu, Weng-Fai Wong arxiv

System prompt optimization improves agent behavior without modifying the underlying model, yielding human-readable, model-agnostic instructions. Existing methods build a prompt agent that refines task agents' system prom…

Code Generation

Swarm Systems as a Platform for Open-Ended Evolutionary Dynamics

2024-09-02 · Hiroki Sayama

Artificial swarm systems have been extensively studied and used in computer science, robotics, engineering and other technological fields, primarily as a platform for implementing robust distributed systems to achieve pr…

Diversity

Group-Evolving Agents: Open-Ended Self-Improvement via Experience Sharing

2026-02-04 · Zhaotian Weng, Antonis Antoniades, Deepak Nathani, Zhen Zhang 외 arxiv

Open-ended self-improving agents can autonomously modify their own structural designs to advance their capabilities and overcome the limits of pre-defined architectures, thus reducing reliance on human intervention. We i…

Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills

2026-07-24 · Siyuan Huang, Pengyu Cheng, Haotian Liu, Tao Chen 외 hf

LLM training is shifting from manual design and annotation to interaction-driven self-evolution. However, existing self-evolutionary methods face a fundamental dilemma between task diversity and verification reliability:…

Reinforcement Learning