学习动力学中的内在-外在耦合
Intrinsic-Extrinsic Coupling in Learning Dynamics
浏览论文内容
中文总结 AI 辅助
本文提出内在-外在耦合框架，通过延续条件干预价值分析学习动力学，证明重放等外部延续与内在干预存在非加性交互，且耦合不必然带来正协同。
中文摘要 AI 辅助
学习者的当前观察并不一定决定其对进一步训练的反应。我们通过受约束的学习状态干预的延续条件价值来表述内在-外在耦合，其中观察相对纤维描述了当前的一致性。一个可执行的有限帧分类器头部写入在有限精度接受检查下保护当前logits，同时修复指定的历史边际。我们区分了局部可容许性、延续条件干预价值和完整策略性能。一个匹配的四单元对比识别了相同内在干预与替代外部延续之间特定于读出的非加性。在CLINC衍生的类增量设置中，重放将写入的32次更新贡献从五个正确预测变为零。在输出蒸馏、RoBERTa骨干网络以及优化器原生的SGDW动力学下，也出现了非零交互。在SGDW下，正确计数交互在128次更新时所有三个激活根均为负，表明耦合不一定意味着正协同。数学分析区分了可行的局部修复和有利的终端输出与训练可达的修复区域。单独的协调测试表明，内容控制匹配或超过开发增益，而一个五根新测试比较（每个臂中都有Fiber存在）显示根依赖而非普遍有益的正确计数效应。在次要的交叉熵读出上，引导分配在所有五对中均产生比标准重放更低的平均损失。总之，这些结果通过将可执行状态几何与延续条件价值、匹配的交互识别和闭环协调联系起来，使内在-外在耦合可操作化，同时将已识别的耦合与完整策略性能区分开来。
英文摘要
A learner's current observations need not determine its response to further training. We formulate intrinsic-extrinsic coupling through the continuation-conditioned value of a constrained learning-state intervention, with observation-relative fibers describing present agreement. An executable finite-frame classifier-head write protects current logits while repairing specified historical margins under finite-precision acceptance checks. We distinguish local admissibility, continuation-conditioned intervention value, and complete-policy performance. A matched four-cell contrast identifies readout-specific non-additivity between the same intrinsic intervention and alternative external continuations. In a CLINC-derived class-incremental setting, replay changes the write's 32-update contribution from five correct predictions to zero. Nonzero interactions also occur under output distillation, with a RoBERTa backbone, and under optimizer-native SGDW dynamics. Under SGDW, correct-count interactions are negative in all three activated roots at 128 updates, showing that coupling need not imply positive synergy. The mathematical analysis distinguishes feasible local repairs and favorable terminal outputs from training-reachable repair regions. Separate coordination tests show that content controls match or exceed the development gain, while a five-root fresh-test comparison with Fiber present in every arm shows root-dependent rather than uniformly beneficial correct-count effects. On the secondary cross-entropy readout, guided allocation yields lower mean loss than standard replay in all five pairs. Together, these results make intrinsic-extrinsic coupling operational by connecting executable state geometry to continuation-conditioned value, matched interaction identification, and closed-loop coordination, while separating identified coupling from complete-policy performance.
