交替任务下可塑性的无量纲控制：从进化生物学到持续学习
Dimensionless Controls of Plasticity Under Alternating Tasks: From Evolutionary Biology to Continual Learning
浏览论文内容
中文总结 AI 辅助
本研究借鉴进化生物学思路，将交替任务下的深度学习可塑性简化为任务分歧度$r$和可达性$ηT$两个无量纲控制量，推导了可塑性边界并通过实验验证了主导因素，得到最优可达性的逆幂律启发式规则。
中文摘要 AI 辅助
变化环境下的可塑性是进化生物学与持续学习领域的核心问题。受近期基因型-表型图谱相关研究的启发，我们研究了一个最简深度学习类比模型：在两个布尔标签集上交替训练网络，并探究哪些生物学可塑性控制机制能迁移到梯度下降场景中。\n将四种已提出的生物学因子重新诠释为训练动力学的量化指标后，我们发现该系统可简化为两个无量纲控制量：任务分歧度$r$（即存在分歧的标签占比），以及可达性$ηT$（学习率与切换周期的乘积）。我们推导出两个可塑性边界：$r$单独决定了乌托邦距离的极值几何下限，而$r$与$ηT$共同约束遗忘程度。\n在9720条训练轨迹上开展的方差分析（ANOVA）证实，$r$、$η$和$T$起主导作用，而（生物学场景中重点关注的）中性集大小的影响可忽略不计。最优可达性本身近似服从逆幂律$ηT^{*}\propto r^{-1.18}$，由此得到一种仅通过任务分歧度就能设定最优可达性$ηT^*$的启发式方法。因此，成立的类比是动力学层面而非几何层面的，我们的研究框架为从物理学和工程学中其他受驱系统的视角研究可塑性提供了可能。
英文摘要
Plasticity under changing environments is central to both evolutionary biology and continual learning. Motivated by recent work on genotype--phenotype maps, we study a minimal deep-learning analogue where a network is trained alternately on two Boolean label sets, and ask which biological controls of plasticity survive the translation to gradient descent. Reinterpreting four proposed biological factors as quantities of training dynamics, we find the system reduces to two dimensionless controls: the task disagreement $r$, the fraction of disagreeing labels, and the reach $ηT$, the product of learning rate and switching period. We derive two bounds on plasticity: $r$ alone fixes an extremal geometric floor on the utopia distance, while $r$ and $ηT$ jointly bound forgetting. Across 9,720 trajectories, an ANOVA confirms that $r$, $η$, and $T$ dominate, while the effect of neutral-set size (emphasized in the biological setting) is negligible. The optimal reach itself follows an approximate inverse power law $ηT^{*}\propto r^{-1.18}$, yielding a heuristic that sets the optimal reach $ηT^*$ from the task disagreement alone. The analogy that survives is therefore dynamical rather than geometric, and our setting enables a view of plasticity through the lens of other driven systems in physics and engineering.
发表机构
- University of Chicago（芝加哥大学）
机构由 AI 辅助整理，请以论文原文为准。
