计算机科学
弹道
推论
合成数据
人工智能
机器学习
生成对抗网络
序列(生物学)
生成模型
国家(计算机科学)
生成语法
数据挖掘
深度学习
算法
物理
天文
生物
遗传学
作者
Nils Henke,Shimon Wonsak,Prasenjit Mitra,Michael Nolting,Nicolas Tempelmeier
标识
DOI:10.1007/978-3-031-33377-4_7
摘要
While the ever-increasing amount of available data has enabled complex machine learning algorithms in various application areas, maintaining data privacy has become more and more critical. This is especially true for mobility data. In nearly all cases, mobility data is personal and therefore the drivers’ privacy needs to be protected. However, mobility data is particularly hard to anonymize, hindering its use in machine learning algorithms to its full potential. In this paper, we address these challenges by generating synthetic vehicle trajectories that are not subject to personal data protection but have the same statistical characteristics as the originals. We present CondTraj-GAN– Conditional Trajectory Generative Adversarial Network. – a novel end-to-end framework to generate entirely synthetic vehicle trajectories. We introduce a specialized training and inference procedure that enables the application of GANs to discrete trajectory data conditioned on their sequence length. We demonstrate the data utility of the synthetic trajectories by comparing their spatial characteristics with the original dataset. Finally, our evaluation shows that CondTraj-GAN reliably outperforms state-of-the-art trajectory generation baselines.
科研通智能强力驱动
Strongly Powered by AbleSci AI