One of key challenges of skeleton-based action recognition (SAR) tasks is the complex nature of human motion patterns. Variations such as performers and viewpoints may impose negative effects to the action recognition accuracy. In this work, we propose the Multi-Localized Sensitive Autoencoder-Attention-LSTM (Multi-LiSAAL) for SAR. The Localized Stochastic Sensitive Autoencoder (LiSSA) encodes both spatial and temporal information, and extracts meaningful features from different parts (four limbs and a trunk) from the skeleton. The LiSSA is trained by minimizing the localized generalization error to enhance the robustness of autoencoders via reducing its sensitivity with respect to small variations in inputs. We apply an attention mechanism to assign different weights to different skeleton parts and focus more on informative sections. Then, a backbone classifier network takes weighted features as inputs to differentiates actions. Experimental results on five public benchmarking datasets show that the Multi-LiSAAL outperforms state-of-the-art methods.
25 February 2022 in IEEE Transactions on Neural Networks and Learning Systems
In this brief, we investigate the problem of incremental learning under data stream with emerging new classes (SENC). In the literature, existing approaches encounter the following problems: 1) yielding high false positive for the new class; i) having long prediction time; and 3) having access to true labels for all instances, which is unrealistic and unacceptable in real-life streaming tasks. Therefore, we propose the k-Nearest Neighbor ENSemble-based method (KNNENS) to handle these problems. The KNNENS is effective to detect the new class and maintains high classification performance for known classes. It is also efficient in terms of run time and does not require true labels of new class instances for model update, which is desired in real-life streaming classification tasks. Experimental results show that the KNNENS achieves the best performance on four benchmark datasets and three real-world data streams in terms of accuracy and F1-measure and has a relatively fast run time compared to four reference methods. Codes are available at https://github.com/Ntriver/KNNENS.
18 September 2020 in IET Generation, Transmission & Distribution
Load forecasting is a complex non‐linear problem with high volatility and uncertainty. This study presents a novel load forecasting method known as deep neural network and historical data augmentation (DNN–HDA). The method utilises HDA to enhance regression by DNN for monthly load forecasting, considering that the historical data to have a high correlation with the corresponding predicted data. To make the best use of the historical data, one year's historical data is combined with the basic features to construct the input vector for a predicted load. In this way, if there is C years' historical data, one predicted load can have C input vectors to create the same number of samples. DNN–HDA increases the number of training samples and enhances the generalisation of the model to reduce the forecasting error. The proposed method is tested on daily peak loads from 2006 to 2015 of Austria, Czech and Italy. Comparisons are made between the proposed method and several state‐of‐the‐art models. DNN–HDA outperforms DNN by 44%, 38% and 63% on the three data sets, respectively.
Music genre recognition (MGR) plays a fundamental role in the context of music indexing and retrieval. Unlike images, music genres consist of immediate characteristics that are highly diversified with abstractions in different levels. However, most representation learning methods for MGR focus on global features and make decisions from features in the same level. To remedy such defects, we intergrate a convolutional neural network (CNN) with NetVLAD and self-attention to capture the local information across levels and learn their long-term dependencies. A meta classifier is used to make the final MGR classification by learning from aggregated high-level features from different local feature coding networks. Experimental results show that the proposed approach yields higher accuracies than other state-of-the-art models on GTZAN, ISMIR2004, and Extended Ballroom dataset.