Search for: All records

As the real-world applications (image segmentation, speech recognition, machine translation, etc.) are increasingly adopting Deep Neural Networks (DNNs), DNN's vulnerabilities in a malicious environment have become an increasingly important research topic in adversarial machine learning. Adversarial machine learning (AML) focuses on exploring vulnerabilities and defensive techniques for machine learning models. Recent work has shown that most adversarial audio generation methods fail to consider audios' temporal dependency (TD) (i.e., adversarial audios exhibit weaker TD than benign audios). As a result, the adversarial audios are easily detectable by examining their TD. Therefore, one area of interest in the audio AML community is to develop a novel attack that evades a TD-based detection model. In this contribution, we revisit the LSTM model for audio transcription and propose a new audio attack algorithm that evades the TD-based detection by explicitly controlling the TD in generated adversarial audios. The experimental results show that the detectability of our adversarial audio is significantly reduced compared to the state-of-the-art audio attack algorithms. Furthermore, experiments also show that our adversarial audios remain nearly indistinguishable from benign audios with only negligible perturbation magnitude. 

The robustness and vulnerability of Deep Neural Networks (DNN) are quickly becoming a critical area of interest since these models are in widespread use across real-world applications (i.e., image and audio analysis, recommendation system, natural language analysis, etc.). A DNN's vulnerability is exploited by an adversary to generate data to attack the model; however, the majority of adversarial data generators have focused on image domains with far fewer work on audio domains. More recently, audio analysis models were shown to be vulnerable to adversarial audio examples (e.g., speech command classification, automatic speech recognition, etc.). Thus, one urgent open problem is to detect adversarial audio reliably. In this contribution, we incorporate a separate and yet related DNN technique to detect adversarial audio, namely model quantization. Then we propose an algorithm to detect adversarial audio by using a DNN's quantization error. Specifically, we demonstrate that adversarial audio typically exhibits a larger activation quantization error than benign audio. The quantization error is measured using character error rates. We use the difference in errors to discriminate adversarial audio. Experiments with three the-state-of-the-art audio attack algorithms against the DeepSpeech model show our detection algorithm achieved high accuracy on the Mozilla dataset. 

Low temperature solution processed planar‐structure perovskite solar cells gain great attention recently, while their power conversions are still lower than that of high temperature mesoporous counterpart. Previous reports are mainly focused on perovskite morphology control and interface engineering to improve performance. Here, this study systematically investigates the effect of precise stoichiometry, especially the PbI₂contents on device performance including efficiency, hysteresis and stability. This study finds that a moderate residual of PbI₂can deliver stable and high efficiency of solar cells without hysteresis, while too much residual PbI₂will lead to serious hysteresis and poor transit stability. Solar cells with the efficiencies of 21.6% in small size (0.0737 cm²) and 20.1% in large size (1 cm²) with moderate residual PbI₂in perovskite layer are obtained. The certificated efficiency for small size shows the efficiency of 20.9%, which is the highest efficiency ever recorded in planar‐structure perovskite solar cells, showing the planar‐structure perovskite solar cells are very promising.