arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2312.09078cs.LGcs.NE
收藏

基于最小最大遗憾的鲁棒决策树构建协同进化算法

Coevolutionary Algorithm for Building Robust Decision Trees under Minimax Regret

Warsaw University of Technology(华沙理工大学);The Ohio State University(俄亥俄州立大学);AGH University of Krakow(AGH科技大学)

Adam Żychowski, Andrew Perrault, Jacek Mańdziuk

AI总结:

针对传统决策树在对抗环境下鲁棒性不足的问题,提出协同进化算法CoEvoRDT,通过决策树与扰动特征的交替进化及混合纳什均衡优化,在20个数据集上的最小最大遗憾等指标优于4种前沿方法。

摘要:

近年来,人们对开发能够抵御对抗性攻击的鲁棒机器学习(ML)模型的兴趣日益浓厚,其中包括应用最广泛、高效且可解释的机器学习算法之一——决策树(DTs)。本文提出了一种新颖的协同进化算法(CoEvoRDT),旨在构建能够在对抗环境下处理高维噪声数据的鲁棒决策树。受传统决策树算法局限性的启发,我们利用自适应协同进化机制,使决策树能够在与受扰动输入数据的交互中进化和学习。CoEvoRDT通过交替进化相互竞争的决策树种群和扰动特征种群,从而构建出具备预期特性的决策树。CoEvoRDT可轻松适配多种目标指标,支持使用定制化的鲁棒性准则,例如最小最大遗憾(minimax regret)。此外,CoEvoRDT能够将其他前沿方法生成的决策树纳入初始种群,并在协同进化过程中对其进行优化,因此具备提升这些方法性能的潜力。受博弈论启发,CoEvoRDT采用混合纳什均衡来提升收敛性。该方法在20个常用数据集上进行了测试,与4种前沿算法相比表现更优:在对抗准确率指标上,它在13个数据集上优于所有对比方法;在最小最大遗憾指标上,它在全部20个数据集上均表现最优。出色的实验结果以及误差度量选择的灵活性,使得CoEvoRDT成为实际应用中构建鲁棒决策树的一种极具前景的方法。

英文摘要:

In recent years, there has been growing interest in developing robust machine learning (ML) models that can withstand adversarial attacks, including one of the most widely adopted, efficient, and interpretable ML algorithms-decision trees (DTs). This paper proposes a novel coevolutionary algorithm (CoEvoRDT) designed to create robust DTs capable of handling noisy high-dimensional data in adversarial contexts. Motivated by the limitations of traditional DT algorithms, we leverage adaptive coevolution to allow DTs to evolve and learn from interactions with perturbed input data. CoEvoRDT alternately evolves competing populations of DTs and perturbed features, enabling construction of DTs with desired properties. CoEvoRDT is easily adaptable to various target metrics, allowing the use of tailored robustness criteria such as minimax regret. Furthermore, CoEvoRDT has potential to improve the results of other state-of-the-art methods by incorporating their outcomes (DTs they produce) into the initial population and optimize them in the process of coevolution. Inspired by the game theory, CoEvoRDT utilizes mixed Nash equilibrium to enhance convergence. The method is tested on 20 popular datasets and shows superior performance compared to 4 state-of-the-art algorithms. It outperformed all competing methods on 13 datasets with adversarial accuracy metrics, and on all 20 considered datasets with minimax regret. Strong experimental results and flexibility in choosing the error measure make CoEvoRDT a promising approach for constructing robust DTs in real-world applications.

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。

↑