Wake-up word detection models are widely used in real life, but suffer from severe performance degradation when encountering adversarial samples. In this paper we discuss the concept of confusing words in adversarial samples. Confusing words are commonly encountered, which are various kinds of words that sound similar to the predefined keywords. To enhance the wake word detection system's robustness against confusing words, we propose several methods to generate the adversarial confusing samples for simulating real confusing words scenarios in which we usually do not have any real confusing samples in the training set. The generated samples include concatenated audio, synthesized data, and partially masked keywords. Moreover, we use a domain embedding concatenated system to improve the performance. Experimental results show that the adversarial samples generated in our approach help improve the system's robustness in both the common scenario and the confusing words scenario. In addition, we release the confusing words testing database called HI-MIA-CW for future research.
Different from large-scale platforms such as Taobao and Amazon, developing CVR models in small-scale recommendation scenarios is more challenging due to the severe Data Distribution Fluctuation (DDF) issue. DDF prevents existing CVR models from being effective since 1) several months of data are needed to train CVR models sufficiently in small scenarios, leading to considerable distribution discrepancy between training and online serving; and 2) e-commerce promotions have much more significant impacts on small scenarios, leading to distribution uncertainty of the upcoming time period. In this work, we propose a novel CVR method named MetaCVR from a perspective of meta learning to address the DDF issue. Firstly, a base CVR model which consists of a Feature Representation Network (FRN) and output layers is elaborately designed and trained sufficiently with samples across months. Then we treat time periods with different data distributions as different occasions and obtain positive and negative prototypes for each occasion using the corresponding samples and the pre-trained FRN. Subsequently, a Distance Metric Network (DMN) is devised to calculate the distance metrics between each sample and all prototypes to facilitate mitigating the distribution uncertainty. At last, we develop an Ensemble Prediction Network (EPN) which incorporates the output of FRN and DMN to make the final CVR prediction. In this stage, we freeze the FRN and train the DMN and EPN with samples from recent time period, therefore effectively easing the distribution discrepancy. To the best of our knowledge, this is the first study of CVR prediction targeting the DDF issue in small-scale recommendation scenarios. Experimental results on real-world datasets validate the superiority of our MetaCVR and online A/B test also shows our model achieves impressive gains of 11.92% on PCVR and 8.64% on GMV.
This paper introduces an online speaker diarization system that can handle long-time audio with low latency. First, a new variant of agglomerative hierarchy clustering is built to cluster the speakers in an online fashion. Then, a speaker embedding graph is proposed. We use this graph to exploit a graph-based reclustering method to further improve the performance. Finally, a label matching algorithm is introduced to generate consistent speaker labels, and we evaluate our system on both DIHARD3 and VoxConverse datasets, which contain long audios with various kinds of scenarios. The experimental results show that our online diarization system outperforms the baseline offline system and has comparable performance to our offline system.
Graph representation learning (GRL) is critical for graph-structured data analysis. However, most of the existing graph neural networks (GNNs) heavily rely on labeling information, which is normally expensive to obtain in the real world. Existing unsupervised GRL methods suffer from certain limitations, such as the heavy reliance on monotone contrastiveness and limited scalability. To overcome the aforementioned problems, in light of the recent advancements in graph contrastive learning, we introduce a novel self-supervised graph representation learning algorithm via Graph Contrastive Adjusted Zooming, namely G-Zoom, to learn node representations by leveraging the proposed adjusted zooming scheme. Specifically, this mechanism enables G-Zoom to explore and extract self-supervision signals from a graph from multiple scales: micro (i.e., node-level), meso (i.e., neighbourhood-level), and macro (i.e., subgraph-level). Firstly, we generate two augmented views of the input graph via two different graph augmentations. Then, we establish three different contrastiveness on the above three scales progressively, from node, neighbouring, to subgraph level, where we maximize the agreement between graph representations across scales. While we can extract valuable clues from a given graph on the micro and macro perspectives, the neighbourhood-level contrastiveness offers G-Zoom the capability of a customizable option based on our adjusted zooming scheme to manually choose an optimal viewpoint that lies between the micro and macro perspectives to better understand the graph data. Additionally, to make our model scalable to large graphs, we employ a parallel graph diffusion approach to decouple model training from the graph size. We have conducted extensive experiments on real-world datasets, and the results demonstrate that our proposed model outperforms state-of-the-art methods consistently.
Federated Learning (FL) allows multiple participating clients to train machine learning models collaboratively by keeping their datasets local and only exchanging model updates. Existing FL protocol designs have been shown to be vulnerable to attacks that aim to compromise data privacy and/or model robustness. Recently proposed defenses focused on ensuring either privacy or robustness, but not both. In this paper, we develop a framework called PRECAD, which simultaneously achieves differential privacy (DP) and enhances robustness against model poisoning attacks with the help of cryptography. Using secure multi-party computation (MPC) techniques (e.g., secret sharing), noise is added to the model updates by the honest-but-curious server(s) (instead of each client) without revealing clients' inputs, which achieves the benefit of centralized DP in terms of providing a better privacy-utility tradeoff than local DP based solutions. Meanwhile, a crypto-aided secure validation protocol is designed to verify that the contribution of model update from each client is bounded without leaking privacy. We show analytically that the noise added to ensure DP also provides enhanced robustness against malicious model submissions. We experimentally demonstrate that our PRECAD framework achieves higher privacy-utility tradeoff and enhances robustness for the trained models.
Recently, the attention mechanism such as squeeze-and-excitation module (SE) and convolutional block attention module (CBAM) has achieved great success in deep learning-based speaker verification system. This paper introduces an alternative effective yet simple one, i.e., simple attention module (SimAM), for speaker verification. The SimAM module is a plug-and-play module without extra modal parameters. In addition, we propose a noisy label detection method to iteratively filter out the data samples with a noisy label from the training data, considering that a large-scale dataset labeled with human annotation or other automated processes may contain noisy labels. Data with the noisy label may over parameterize a deep neural network (DNN) and result in a performance drop due to the memorization effect of the DNN. Experiments are conducted on VoxCeleb dataset. The speaker verification model with SimAM achieves the 0.675% equal error rate (EER) on VoxCeleb1 original test trials. Our proposed iterative noisy label detection method further reduces the EER to 0.643%.
This paper managed to induce probability theory (sigma system) and possibility theory (max system) respectively from randomness and fuzziness, through which the premature theory of possibility is expected to be well founded. Such an objective is achieved by addressing three open key issues: a) the lack of clear mathematical definitions of randomness and fuzziness; b) the lack of intuitive mathematical definition of possibility; c) the lack of abstraction procedure of the axiomatic definitions of probability/possibility from their intuitive definitions. Especially, the last issue involves the question why the key axiom of "maxitivity" is adopted for possibility measure. By taking advantage of properties of the well-defined randomness and fuzziness, we derived the important conclusion that "max" is the only but un-strict disjunctive operator that is applicable across the fuzzy event space, and is an exact operator for fuzzy feature extraction that assures the max inference is an exact mechanism. It is fair to claim that the long-standing problem of lack of consensus to the foundation of possibility theory is well resolved, which would facilitate wider adoption of possibility theory in practice and promote cross prosperity of the two uncertainty theories of probability and possibility.
With the development of deep learning, automatic speaker verification has made considerable progress over the past few years. However, to design a lightweight and robust system with limited computational resources is still a challenging problem. Traditionally, a speaker verification system is symmetrical, indicating that the same embedding extraction model is applied for both enrollment and verification in inference. In this paper, we come up with an innovative asymmetric structure, which takes the large-scale ECAPA-TDNN model for enrollment and the small-scale ECAPA-TDNNLite model for verification. As a symmetrical system, our proposed ECAPA-TDNNLite model achieves an EER of 3.07% on the Voxceleb1 original test set with only 11.6M FLOPS. Moreover, the asymmetric structure further reduces the EER to 2.31%, without increasing any computational costs during verification.
Computer vision is widely used in the fields of driverless, face recognition and 3D reconstruction as a technology to help or replace human eye perception images or multidimensional data through computers. Nowadays, with the development and application of deep neural networks, the models of deep neural networks proposed for computer vision are becoming more and more abundant, and developers will use the already trained models on the way to solve problems, and need to consult the relevant documents to understand the use of the model. The class model, which creates the need to quickly and accurately find the relevant models that you need. The automatic annotation method of visual depth neural network proposed in this paper is based on natural language processing technology such as semantic analysis, which realizes automatic labeling of model application fields. In the three top international conferences on computer vision: ICCV, CVPR and ECCV, the average correct rate of application of the papers of 72 papers reached 90%, indicating the effectiveness of the automatic labeling system.
Optimally extracting the advantages available from reconfigurable intelligent surfaces (RISs) in wireless communications systems requires estimation of the channels to and from the RIS. The process of determining these channels is complicated by the fact that the RIS is typically composed of passive elements without any data processing capabilities, and thus the channels must be estimated indirectly by a non-colocated device, typically a controlling base station. In this article, we examine channel estimation for RIS-based systems from a fundamental viewpoint. We study various possible channel models and the identifiability of the models as a function of the available pilot data and behavior of the RIS during training. In particular, we consider situations with and without line-of-sight propagation, single- and multiple-antenna configurations for the users and base station, correlated and sparse channel models, single-carrier and wideband OFDM scenarios, availability of direct links between the users and base station, exploitation of prior information, as well as a number of other special cases. We further conduct numerical comparisons of achievable performance for various channel models using the relevant Cramer-Rao bounds.