首页/论文/2003.06259

arXiv 2003.062592020-03-13cs.LGstat.ML

Taylor Expansion Policy Optimization

Columbia University（哥伦比亚大学）；Google DeepMind（谷歌DeepMind）

Yunhao Tang, Michal Valko, Rémi Munos

英文摘要：

In this work, we investigate the application of Taylor expansions in reinforcement learning. In particular, we propose Taylor expansion policy optimization, a policy optimization formalism that generalizes prior work (e.g., TRPO) as a first-order special case. We also show that Taylor expansions intimately relate to off-policy evaluation. Finally, we show that this new formulation entails modifications which improve the performance of several state-of-the-art distributed algorithms.

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。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

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。

继续探索

相关论文

折扣因子的泰勒展开

Taylor Expansion of Discount Factors

Columbia University（哥伦比亚大学）；DeepMind（深度思维）

本文研究强化学习中折扣因子不一致的影响，提出插值目标函数族，并据此改进价值函数估计与策略优化，带来性能提升及对深度RL算法的深入理解。

2026-10-06cs.LGstat.ML

TaylorPODA：一种基于泰勒展开的方法，用于改进不透明模型的事后归因

TaylorPODA: A Taylor Expansion-Based Method to Improve Post-Hoc Attributions for Opaque Models

研究针对不透明模型事后归因方法缺乏普遍标准的问题，提出基于泰勒展开框架的TaylorPODA方法，通过引入假设、分析矛盾、设计可控分配机制等，使其既满足假设又适应下游目标，提升了归因与用户定义效用的对齐及解释的可交流性。

2026-07-29cs.AIcs.LGstat.ML

TEAM: An Taylor Expansion-Based Method for Generating Adversarial Examples

School of Big-Data Science, Zhejiang University of Science and Technology（浙江科技学院大数据学院）；Network and Information Security Laboratory of Hangzhou Hikvision Digital Technology Co., Ltd.（杭州海康威视数字技术股份有限公司网络与信息安全实验室）；College of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院）；School of Information and Electronic Engineering, Zhejiang University of Science and Technology（浙江科技学院信息与电子工程学院）；Laboratoire ETIS UMR-CNRS 8051, CY Cergy Paris Université（CY塞尔吉巴黎大学ETIS实验室（法国国家科学研究中心联合研究单位））

Although Deep Neural Networks(DNNs) have achieved successful applications in many fields, they are vulnerable to adversarial examples.Adversarial training is one of the most effective methods to improve the robustness of DNNs, and it is generally considered as solving a saddle point problem that minimizes risk and maximizes perturbation.Therefore, powerful adversarial examples can effectively replicate the situation of perturbation maximization to solve the saddle point problem.The method proposed in this paper approximates the output of DNNs in the input neighborhood by using the Taylor expan…

2020-03-25cs.LGstat.ML

基于高斯过程的概率泰勒展开

A probabilistic Taylor expansion with Gaussian processes

University of Helsinki（赫尔辛基大学）；University of Southampton（南安普顿大学）；Lund University（隆德大学）；Aalto University（阿尔托大学）

该研究提出一类基于Taylor核的高斯过程框架，其特定数据下的后验均值可复现任意阶截断泰勒展开，还证明了Taylor核参数最大似然估计的相关结果，是再生核希尔伯特空间正交数据下高斯过程回归的特例。

2023-08-28cs.LGcs.NAmath.NA

基于泰勒展开的Kolmogorov-Arnold网络用于盲图像质量评估

Taylor expansion-based Kolmogorov-Arnold network for blind image quality assessment

Communication University of China（通信大学）

针对KAN在盲图像质量评估中处理高维特征时性能受限且计算成本高的问题，提出TaylorKAN，利用泰勒展开作为可学习激活函数增强局部逼近能力，并结合网络深度缩减和特征维度压缩提升计算效率，在五个数据库上表现优于其他KAN模型。

2025-05-27cs.CVcs.LGeess.IV

统计泰勒展开：一种新的、路径无关的不确定性分析方法

Statistical Taylor Expansion: A New and Path-Independent Method for Uncertainty Analysis

Grossman Street, Melville, NY 11747, USA（美国纽约州梅尔维尤市格罗斯曼街11747号）

本文提出了一种新的路径无关的不确定性分析方法，通过将精确输入变量替换为具有已知分布和样本数的随机变量，计算每个结果的均值、偏差和可靠因子，从而实现对输入不确定性的传播追踪，使最终结果成为路径无关的，与传统数学方法不同。

2026-10-19cs.LGstat.CO

更多相关论文收起更多论文

Taylor Unswift：基于泰勒展开的大语言模型权重安全发布方法

Taylor Unswift: Secured Weight Release for Large Language Models via Taylor Expansion

Rice University（莱斯大学）；Rutgers University（罗格斯大学）；Texas A&M University（德克萨斯农工大学）；Case Western Reserve University（凯斯西储大学）；University of Minnesota（明尼苏达大学）

本文提出 TaylorMLP，通过将大语言模型权重转换为泰勒级数参数来保护模型所有权，并利用增加泰勒级数项数降低生成速度以防止大规模滥用，实验表明其可使延迟增加超过 4 倍且输出与原始模型精确匹配。

2025-03-11cs.AIcs.CLcs.CR

Efficient Meta-Learning for Continual Learning with Taylor Expansion Approximation

Boston University（波士顿大学）；Peking University（北京大学）

Continual learning aims to alleviate catastrophic forgetting when handling consecutive tasks under non-stationary distributions. Gradient-based meta-learning algorithms have shown the capability to implicitly solve the transfer-interference trade-off problem between different examples. However, they still suffer from the catastrophic forgetting problem in the setting of continual learning, since the past data of previous tasks are no longer available. In this work, we propose a novel efficient meta-learning algorithm for solving the online continual learning problem, where the regularization t…

2022-10-03cs.LG

TANTE：基于神经泰勒展开的时间自适应算子学习

TANTE: Time-Adaptive Operator Learning via Neural Taylor Expansion

Yale University（耶鲁大学）

针对时间相关偏微分方程算子学习中固定步长导致误差累积的问题，提出TANTE框架，通过神经泰勒展开实现自适应步长连续预测，在多基准上实现精度与推理速度的显著提升。

2025-10-06cs.AIcs.LG

Randomized Adversarial Training via Taylor Expansion

Institute of Software, CAS（中国科学院软件研究所）；University of Liverpool（利物浦大学）；Lancaster University（兰卡斯特大学）

In recent years, there has been an explosion of research into developing more robust deep neural networks against adversarial examples. Adversarial training appears as one of the most successful methods. To deal with both the robustness against adversarial examples and the accuracy over clean examples, many works develop enhanced adversarial training methods to achieve various trade-offs between them. Leveraging over the studies that smoothed update on weights during training may help find flat minima and improve generalization, we suggest reconciling the robustness-accuracy trade-off from ano…

2023-03-19cs.AIcs.LG