1 Introduction Time seriesaugmentationis an essential approachto solvethe overfitting problem on the time series classification(TSC)task[1,2].Although existing approaches perform better in mitigating this problem,none...
详细信息
1 Introduction Time seriesaugmentationis an essential approachto solvethe overfitting problem on the time series classification(TSC)task[1,2].Although existing approaches perform better in mitigating this problem,none of them focus on protecting saliency regions on time *** key informative shapelets contained in these regions are the core basis for distinguishing categories(e.g.,upward spikes in ECG and high amplitude in Sensor).
In recent years, social enterprises have gradually emerged as a major force in community governance. In order for social enterprises to better participate in community governance, this paper examines the trust mechani...
详细信息
Visible watermark removal which involves watermark cleaning and background content restoration is pivotal to evaluate the resilience of watermarks. Existing deep neural network (DNN)-based models still struggle with l...
详细信息
Forecasting stock prices is difficult because of the many unknowns and diverse factors that affect the financial market. Using time series data, the study attempts to assess how well the ARIMA (Auto Regressive Integra...
详细信息
Both structured market data and unstructured financial news data can significantly affect stock price fluctuations. Therefore, relying only on a single data source for stock price trend prediction may produce informat...
详细信息
Both structured market data and unstructured financial news data can significantly affect stock price fluctuations. Therefore, relying only on a single data source for stock price trend prediction may produce information bias. This article combines these two types of heterogeneous data and systematically compares the effectiveness of different data fusion technologies to establish the optimal data integration strategy, thereby improving the accuracy of stock price prediction. Specifically, this paper innovatively proposes a stock price trend prediction model named MVL-SVM. The model successfully combines multi-perspective learning with support vector machine, and realizes the effective fusion of stock market data and financial news data. It is worth mentioning that the prediction accuracy of MVL-SVM model using direct fusion method is more than 30% higher than that of index modeling method, which indicates that direct fusion method can fully learn effective information from multiple information sources, thus significantly improving the prediction effect.
Vehicle movement is frequently captured in the form of GPS trajectories, i.e., sequences of timestamped GPS locations. Such data is widely used for various tasks such as travel-time estimation, trajectory recovery, an...
详细信息
We consider how to make dynamic pricing decision for Chinese Online (COL) at T time-points, an online publisher that allow authors to sell their ongoing book projects. Instead of paying for a book, readers pay for eac...
详细信息
Efficient energy data analysis and then prediction can be especially beneficial in the Indian Subcontinent with its ever-growing need and industrial expansion. The authors developed and compared three different machin...
详细信息
Current speaker recognition systems primarily rely on supervised approaches, constrained by the scale of labeled datasets. To boost the system performance, researchers leverage large pretrained models such as WavLM to...
Current speaker recognition systems primarily rely on supervised approaches, constrained by the scale of labeled datasets. To boost the system performance, researchers leverage large pretrained models such as WavLM to transfer learned high-level features to the downstream speaker recognition task. However, this approach introduces extra parameters as the pretrained model remains in the inference stage. Another group of researchers directly apply self-supervised methods such as DINO to speaker embedding learning, yet they have not explored its potential on large-scale in-the-wild datasets. In this paper, we present the effectiveness of DINO training on the large-scale WenetSpeech dataset and its transferability in enhancing the supervised system performance on the CNCeleb dataset. Additionally, we introduce a confidence-based data filtering algorithm to remove unreliable data from the pretraining dataset, leading to better performance with less training data. The associated pretrained models, confidence files, pretraining and finetuning scripts will be made available in the Wespeaker toolkit.
Unsupervised domain adaptation excels in transferring knowledge from a labeled source domain to an unlabeled target domain, playing a critical role in time series applications. Existing time series domain adaptation m...
ISBN:
(纸本)9798331314385
Unsupervised domain adaptation excels in transferring knowledge from a labeled source domain to an unlabeled target domain, playing a critical role in time series applications. Existing time series domain adaptation methods either ignore frequency features or treat temporal and frequency features equally, which makes it challenging to fully exploit the advantages of both types of features. In this paper, we delve into transferability and discriminability, two crucial properties in transferable representation learning. It's insightful to note that frequency features are more discriminative within a specific domain, while temporal features show better transferability across domains. Based on the findings, we propose Adversarial CO-learning Networks (ACON), to enhance transferable representation learning through a collaborative learning manner in three aspects: (1) Considering the multi-periodicity in time series, multi-period frequency feature learning is proposed to enhance the discriminability of frequency features; (2) Temporal-frequency domain mutual learning is proposed to enhance the discriminability of temporal features in the source domain and improve the transferability of frequency features in the target domain; (3) Domain adversarial learning is conducted in the correlation subspaces of temporal-frequency features instead of original feature spaces to further enhance the transferability of both features. Extensive experiments conducted on a wide range of time series datasets and five common applications demonstrate the state-of-the-art performance of ACON. Code is available at https://***/mingyangliu1024/ACON.
暂无评论