论文/arXiv 2610.11662·20 天前

Lamarck的驾驶学校：通过进化竞争发现自动驾驶训练策略

针对自动驾驶训练场景分布价值难以由预设指标刻画的问题，提出Lamarckian进化框架，并在实验中验证。

AI 整理

• 本体 SDK 与控制
• 具身模型

原文链接：Lamarck's Driving School: Discovering Autonomous Driving Training Strategies through Evolutionary Competition（arXiv 2610.11662 · 发布日期：2026-10-08）

Lamarck's Driving School: Discovering Autonomous Driving Training Strategies through Evolutionary Competition

针对自动驾驶训练场景分布价值难以由预设指标刻画的问题，提出Lamarckian进化框架，并在实验中验证。

核心要点

• 自动驾驶能力强依赖训练场景分布，现实性、难度或风险等代理指标可能误判训练价值。
• 将训练策略发现表述为多阶段双层优化，用Lamarckian进化算法近似求解。
• 外层用交叉、变异和选择探索分布，内层学习策略并继承能力，使二者共同演化。
• 完整框架最高降25.07%性能损失，轻量策略降19.13%；代码在GitHub发布。

英文摘要 (Abstract)

Autonomous driving capabilities depend strongly on the distribution of scenarios encountered during training. Existing methods commonly construct or dynamically adapt training scenario distributions using surrogate criteria such as realism, difficulty, or risk. However, these predefined surrogates may misrepresent training value, leading to inefficient use of training resources. To address this limitation, we propose a Lamarckian evolutionary framework that replaces surrogate-based guidance with competition among candidate distributions. We formulate training strategy discovery as a multi-stage bilevel optimization problem and use Lamarckian evolution algorithm to approximate its solution. At the outer level, Darwinian crossover, mutation, and selection explore the scenario distribution space; at the inner level, policy learning acquires new capabilities, and Lamarckian inheritance transfers them to subsequent stages, allowing scenario distributions and policy capabilities to co-evolve. The resulting evolutionary trajectories reveal recurring stage-wise regularities among high-value distributions, characterized by capability accumulation through stage-wise challenge rotation. We further distill these regularities into a lightweight, reusable Lamarckian Training Strategy. Experiments show that, compared with the baseline, the complete framework reduces performance loss by up to 25.07%, while the lightweight strategy still achieves a 19.13% reduction. These results demonstrate that evolutionary competition can both discover effective training strategies and reveal reusable stage-wise patterns in how the value of training distributions changes with policy capability. Code is available on GitHub.

作者：Yichun Ye, He Zhang, Ye Tian, Jian Sun

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.11662：Lamarck's Driving School: Discovering Autonomous Driving Training Strategies through Evolutionary Competition访问 2026-10-21