paper-with-me

Papers

DATTA: Domain-Adversarial Test-Time Adaptation for Cross-Domain WiFi-Based Human Activity Recognition

2024-11-20 · Julian Strohmayer, Rafael Sterzinger, Matthias Wödlinger, Martin Kampel

Cross-domain generalization is an open problem in WiFi-based sensing due to variations in environments, devices, and subjects, causing domain shifts in channel state information. To address this, we propose Domain-Adversarial Test-Time Adaptation (DATTA), a novel framework combining domain-adversarial training (DAT), test-time adaptation (TTA), and weight resetting to facilitate adaptation to unseen target domains and to prevent catastrophic forgetting. DATTA is integrated into a lightweight, flexible architecture optimized for speed. We conduct a comprehensive evaluation of DATTA, including an ablation study on all key components using publicly available data, and verify its suitability for real-time applications such as human activity recognition. When combining a SotA video-based variant of TTA with WiFi-based DAT and comparing it to DATTA, our method achieves an 8.1% higher F1-Score. The PyTorch implementation of DATTA is publicly available at: https://github.com/StrohmayerJ/DATTA.

📄 PDF Abstract BibTeX arXiv:2411.13284

Code (1)

strohmayerj/datta 공식 구현 pytorch

Tasks

Activity RecognitionDomain GeneralizationHuman Activity RecognitionTest-time Adaptation

Similar Papers 제목 키워드 기반

DATTA: Towards Diversity Adaptive Test-Time Adaptation in Dynamic Wild World

2024-08-15 · Chuyang Ye, Dongyan Wei, Zhendong Liu, Yuanyi Pang 외

Test-time adaptation (TTA) effectively addresses distribution shifts between training and testing data by adjusting models on test samples, which is crucial for improving model inference in real-world applications. Howev…

DiversityTest-time Adaptation

QuadAttac$K$: A Quadratic Programming Approach to Learning Ordered Top-$K$ Adversarial Attacks

2023-09-21 · NeurIPS 2023 11

The adversarial vulnerability of Deep Neural Networks (DNNs) has been well-known and widely concerned, often under the context of learning top-$1$ attacks (e.g., fooling a DNN to classify a cat image as dog). This paper …

QuadAttack: A Quadratic Programming Approach to Ordered Top-K Attacks

2023-12-12 · Thomas Paniagua, Ryan Grainger, Tianfu Wu

The adversarial vulnerability of Deep Neural Networks (DNNs) has been well-known and widely concerned, often under the context of learning top-$1$ attacks (e.g., fooling a DNN to classify a cat image as dog). This paper …

LiDAttack: Robust Black-box Attack on LiDAR-based Object Detection

2024-11-04 · Jinyin Chen, Danxin Liao, Sheng Xiang, Haibin Zheng

Since DNN is vulnerable to carefully crafted adversarial examples, adversarial attack on LiDAR sensors have been extensively studied. We introduce a robust black-box attack dubbed LiDAttack. It utilizes a genetic algorit…

Adversarial Attackobject-detectionObject Detection

MedAttacker: Exploring Black-Box Adversarial Attacks on Risk Prediction Models in Healthcare

2021-12-11 · Muchao Ye, Junyu Luo, Guanjie Zheng, Cao Xiao 외

Deep neural networks (DNNs) have been broadly adopted in health risk prediction to provide healthcare diagnoses and treatments. To evaluate their robustness, existing research conducts adversarial attacks in the white/gr…

Adversarial AttackPositionReinforcement Learning (RL)