In this paper, we propose a method for defending against audio adversarial examples that operates by applying audio style transfer learning. The proposed method has the effect of maintaining the classification result produced by the target model and removing the adversarial noise by changing only the style while maintaining the content of the input audio sample. In an experimental evaluation using the Mozilla Common Voice dataset as the test data source and TensorFlow as the machine learning library, the proposed method improved the target model’s accuracy on the adversarial examples from 2.1% to 79.2% while maintaining its accuracy on the original samples at 81.4%.
A smart regenerative braking system for EVs can reduce unnecessary brake operations by assisting in the braking of a vehicle according to the driving situation, road slope, and driver’s preference. Since the strength of regenerative braking is generally determined based on calibration data determined during the vehicle development process, some drivers could encounter inconveniences when the regenerative braking is activated differently from their driving habits. In order to solve this problem, various deep learning-based algorithms have been developed to provide driving stability by learning the driving data. Among those artificial intelligence algorithms, anomaly detection algorithms can successfully separate the deceleration data in abnormal driving situations, and the resulting refined deceleration data can be used to train the regression model to achieve better driving stability. This study evaluates the performance of a personalized driving assistance system by applying driver characteristic data, obtained through an anomaly detection algorithm, to vehicle control.
30 November 2022 in The Journal of Korean Institute of Information Technology
이동 경로 데이터는 각종 모바일 디바이스를 이용하여 GPS 위치를 연속적으로 갖는 데이터를 의미하며, 이러한 이동 경로 데이터가 기계 학습에 활용되기 위해서는 데이터 검수 및 정제가 중요하다. 이 중 이상치 탐지 기술은 데이터셋의 이상 데이터를 찾아 데이터의 품질을 향상시켜 기계 학습의 성능을 향상시킨다. 본 논문은 사용자 이동 경로에 대한 이상치를 유형별로 기술하고, 5가지 딥러닝 모델(CNN, DNN, LSTM, Bi-LSTM, LSTM 오토인코더)을 사용하여 이상치를 탐지하는 모델을 제안한다. 또한, 5가지 딥러닝 모델 중 가장 적합한 모델을 선정하기 위하여 실제 이동 경로 데이터셋을 각 모델에 학습시키고 이상치를 탐지하여 그 성능을 측정한다. 비교 평가 결과, LSTM 오토인코더 모델이 다른 모델에 비해 이상치 탐지를 위한 F1-score가 가장 높았으며 안정적인 성능을 보였다.
For electric vehicles, the Adaptive Cruise Control (ACC) in Advanced Driver Assistance Systems (ADAS) is designed to assist braking based on driving conditions, road inclines, predefined deceleration strengths, and user braking patterns. However, the driving data collected during the development of ADAS are generally limited and lack diversity. This deficiency leads to late or aggressive braking for different users. Crucially, it is necessary to effectively identify anomalies, such as unexpected or inconsistent braking patterns in ADAS, especially given the challenge of working with unlabelled, limited, and noisy datasets from real-world electric vehicles. In order to tackle the aforementioned challenges in ADAS, we propose Graph Neural Controlled Differential Equation Normalizing Flow (GDFlow), a model that leverages Normalizing Flow (NF) with Neural Controlled Differential Equations (NCDE) to learn the distribution of normal driving patterns continuously. Compared to the traditional clustering or anomaly detection algorithms, our approach effectively captures the spatio-temporal information from different sensor data and more accurately models continuous changes in driving patterns. Additionally, we introduce a quantile-based maximum likelihood objective to improve the likelihood estimate of the normal data near the boundary of the distribution, enhancing the model's ability to distinguish between normal and anomalous patterns. We validate GDFlow using real-world electric vehicle driving data that we collected from Hyundai IONIQ5 and GV80EV, achieving state-of-the-art performance compared to six baselines across four dataset configurations of different vehicle types and drivers. Furthermore, our model outperforms the latest anomaly detection methods across four time series benchmark datasets. Our approach demonstrates superior efficiency in inference time compared to existing methods.