果蝇全连接组网络可以学习人类设计的认知任务
A Drosophila Whole-Connectome Network Can Learn Human-Designed Cognitive Tasks
浏览论文内容
中文总结 AI 辅助
本研究利用果蝇全连接组作为固定循环拓扑，训练有界加法和接地关系语言模型，发现解剖结构提供可复用归纳偏置，显著优于随机重连。
中文摘要 AI 辅助
一个生物接线图能否在其进化而来的行为之外，作为有用的计算基底？我们使用公开的MaleCNS v1.0连接组，该连接组由单个成年雄性果蝇标本重建而来，作为人工网络的固定循环拓扑。我们分别训练了有界加法和一个基于固定100词词汇表的受控接地关系语言任务的模型。在两个模型中，每条解剖边学习一个标量。解剖图在保留的加法测试上达到92.77%的平均准确率，而定向度保持重连的准确率为67.93%。在严格的配对语言端点上，该端点将原始和顺序反转的场景与其对应描述匹配，在四个固定接口上达到61.59%，而匹配重连的准确率为44.17%。在规范接口上，它在固定的21图比较中排名第一。在匹配的48组干预子集上，打乱任务定义的感官特征将其得分从60.94%降至19.27%。这些结果共同表明，高阶MaleCNS接线为有界加法和接地关系语言提供了可复用的归纳偏置。
英文摘要
Can a biological wiring diagram serve as a useful computational substrate beyond the behaviors for which it evolved? We use the publicly released MaleCNS v1.0 connectome, reconstructed from a single adult male Drosophila specimen, as the fixed recurrent topology of an artificial network. We train separate models for bounded addition and for a controlled grounded relational language task built from a fixed 100-word lexicon. In both models, one scalar is learned per anatomical edge. The anatomical graph reaches 92.77% mean accuracy on held-out addition, compared with 67.93% for directed degree-preserving rewires. On the strict paired language endpoint, which matches original and order-reversed scenes to their corresponding descriptions, it reaches 61.59% across four fixed interfaces, compared with 44.17% for matched rewires. At the canonical interface, it ranks first in a fixed 21-graph comparison. On the matched 48-group intervention subset, shuffling task-defined sensory features reduces its score from 60.94% to 19.27%. Together, these results show that higher-order MaleCNS wiring provides a reusable inductive bias for bounded addition and grounded relational language.
发表机构
- Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）
机构由 AI 辅助整理，请以论文原文为准。
