arXiv cs.AI· Andrew W. Lo, Chaoyi Zhao·· 11 小时前AI 评分31
AI 自我进化框架：达尔文、拉马克与智能体导向演化
Adapting the Process of Adaptation: Darwinian, Lamarckian, and Agent-Directed Evolution in AI
AI 导读
一项 arXiv 理论研究提出框架，比较达尔文演化、拉马克演化与智能体导向演化在环境变化后的适应速度：学习足够有效时三者依次更快，生成更多行为变体可加速适应，最优搜索投入取决于学习能力、问题难度与实验成本。但智能体自我导向改进并不保证按预期目标评估，评估目标设定错误时更多搜索会加速收敛到错误行为，递归自我改进中此类错误可能跨代延续，需独立评估与监督同步跟进。
正文 · 原文
Abstract:Evolution has shaped intelligence for billions of years. Artificial intelligence (AI) now raises a striking possibility: intelligence may begin to direct its own evolution. As AI agents become capable of modifying and evaluating their own behavior, how quickly could they improve, and how can that improvement remain safe? We develop a theoretical framework comparing Darwinian, Lamarckian, and agent-directed evolution following environmental change. Darwinian evolution relies on natural selection, Lamarckian evolution adds inherited learning, and agent-directed evolution allows agents to generate and evaluate behavioral variants and choose how extensively to search. When learning is sufficiently effective, these mechanisms exhibit progressively faster adaptation. Generating more variants accelerates adaptation, while optimal search effort depends on learning ability, problem difficulty, and experimentation costs. However, an agent's ability to direct its improvement does not ensure that improvements are evaluated against the intended objective. When the evaluation target is misspecified, additional search accelerates convergence toward the wrong behavior, even as measured performance improves. In recursive self-improvement, such errors may persist across generations, posing challenges for AI safety. Independent evaluation and oversight must keep pace with autonomous AI development. The central challenge is not simply to accelerate the evolution of intelligence, but to preserve humanity's ability to guide its direction, limit its speed, and intervene when necessary.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.07048 [cs.AI] |
| (or arXiv:2611.07048v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.07048 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Andrew Lo [view email]
[v1]
Wed, 4 Nov 2026 19:08:46 UTC (122 KB)
来源：arXiv cs.AI · arxiv.org