arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.04557cs.LGmath.DSmath.OCq-fin.CP
收藏

面向核回归的遗憾最优联邦迁移学习及其在美式期权定价中的应用

Regret-Optimal Federated Transfer Learning for Kernel Regression with Applications in American Option Pricing

McMaster University(麦克马斯特大学);Vector Institute(矢量研究所);ETH Zürich(苏黎世联邦理工学院);University of Basel(巴塞尔大学)

Xuwei Yang, Anastasis Kratsios, Florian Krach, Matheus Grasselli, Aurelien Lucchi

AI总结:

本文针对联邦迁移学习提出遗憾最优迭代方案,在有限秩核回归下推导显式更新,并开发近遗憾最优启发式算法以降低计算复杂度,同时证明其对抗鲁棒性,并通过美式期权定价数值实验验证理论结果。

摘要:

我们提出一种用于联邦迁移学习的最优迭代方案,其中中央规划者可访问用于同一学习模型 $f_θ$ 的数据集 ${\cal D}_1,\dots,{\cal D}_N$。我们的目标是在遵守算法停止时产生的模型 $f_{θ(T)}$ 的损失函数的同时,最小化所有 $T$ 次迭代中生成的参数 $\{θ_i(t)\}_{t=0}^T$ 相对于针对每个数据集获得的专用参数 $θ^\star_{1},\ldots,θ^\star_N$ 的累积偏差。我们只允许每个专用模型(节点/智能体)与中央规划者(服务器)之间在每次迭代(轮次)进行持续通信。针对模型 $f_θ$ 为有限秩核回归的情形,我们推导出遗憾最优算法的显式更新。通过利用遗憾最优算法中的对称性,我们进一步开发了一种近遗憾最优的启发式方法,其运行所需的初等运算减少 $\mathcal{O}(Np^2)$,其中 $p$ 是参数空间的维度。此外,我们研究了遗憾最优算法的对抗鲁棒性,表明一个在所有训练集上对 $q$ 个训练对施加至多 $\varepsilon>0$ 扰动的对手,最多只能将遗憾最优算法的遗憾减少 $\mathcal{O}(\varepsilon q \bar{N}^{1/2})$,其中 $\bar{N}$ 是训练对的总数。为验证我们的理论发现,我们在美式期权定价背景下利用随机生成的有限秩核进行了数值实验。

英文摘要:

We propose an optimal iterative scheme for federated transfer learning, where a central planner has access to datasets ${\cal D}_1,\dots,{\cal D}_N$ for the same learning model $f_θ$. Our objective is to minimize the cumulative deviation of the generated parameters $\{θ_i(t)\}_{t=0}^T$ across all $T$ iterations from the specialized parameters $θ^\star_{1},\ldots,θ^\star_N$ obtained for each dataset, while respecting the loss function for the model $f_{θ(T)}$ produced by the algorithm upon halting. We only allow for continual communication between each of the specialized models (nodes/agents) and the central planner (server), at each iteration (round). For the case where the model $f_θ$ is a finite-rank kernel regression, we derive explicit updates for the regret-optimal algorithm. By leveraging symmetries within the regret-optimal algorithm, we further develop a nearly regret-optimal heuristic that runs with $\mathcal{O}(Np^2)$ fewer elementary operations, where $p$ is the dimension of the parameter space. Additionally, we investigate the adversarial robustness of the regret-optimal algorithm showing that an adversary which perturbs $q$ training pairs by at-most $\varepsilon>0$, across all training sets, cannot reduce the regret-optimal algorithm's regret by more than $\mathcal{O}(\varepsilon q \bar{N}^{1/2})$, where $\bar{N}$ is the aggregate number of training pairs. To validate our theoretical findings, we conduct numerical experiments in the context of American option pricing, utilizing a randomly generated finite-rank kernel.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑