Tax evasion detection plays a crucial role in addressing tax revenue loss. Many efforts have been made to develop tax evasion detection models by leveraging machine learning techniques, but they have not constructed a uniform model for different geographical regions because an ample supply of training examples is a fundamental prerequisite for an effective detection model. When sufficient tax data are not readily available, the development of a representative detection model is more difficult due to unequal feature distributions in different regions. Existing methods face a challenge in explaining and tracing derived results. To overcome these challenges, we propose an Inter-Region Tax Evasion Detection method based on Transfer Learning (IRTED-TL), which is optimized to simultaneously augment training data and induce interpretability into the detection model. We exploit evasion-related knowledge in one region and leverage transfer learning techniques to reinforce the tax evasion detection tasks of other regions in which training examples are lacking. We provide a unified framework that takes advantage of auxiliary data using a transfer learning mechanism and builds an interpretable classifier for inter-region tax evasion detection. Experimental tests based on real-world tax data demonstrate that the IRTED-TL can detect tax evaders with higher accuracy and better interpretability than existing methods.
Adversarial patch is one of the important forms of performing adversarial attacks in the physical world. To improve the naturalness and aggressiveness of existing adversarial patches, location-aware patches are proposed, where the patch's location on the target object is integrated into the optimization process to perform attacks. Although it is effective, efficiently finding the optimal location for placing the patches is challenging, especially under the black-box attack settings. In this paper, we propose the Distribution-Optimized Adversarial Patch (DOPatch), a novel method that optimizes a multimodal distribution of adversarial locations instead of individual ones. DOPatch has several benefits: Firstly, we find that the locations' distributions across different models are pretty similar, and thus we can achieve efficient query-based attacks to unseen models using a distributional prior optimized on a surrogate model. Secondly, DOPatch can generate diverse adversarial samples by characterizing the distribution of adversarial locations. Thus we can improve the model's robustness to location-aware patches via carefully designed Distributional-Modeling Adversarial Training (DOP-DMAT). We evaluate DOPatch on various face recognition and image recognition tasks and demonstrate its superiority and efficiency over existing methods. We also conduct extensive ablation studies and analyses to validate the effectiveness of our method and provide insights into the distribution of adversarial locations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.