Deep learning-based human activity recognition (HAR) methods have shown great promise in the applications of smart healthcare systems and wireless body sensor network (BSN). Despite their demonstrated performance in laboratory settings, the real-world implementation of such methods is still hindered by the cross-subject issue when adapting to new users. To solve this issue, we propose ActiveSelfHAR, a framework that combines active learning's benefit of sparsely acquiring data with actual labels and self- training's benefit of effectively utilizing unlabeled data to enable the deep model to adapt to the target domain, i.e., the new users. In this framework, the model trained in the last iteration or the source domain is first utilized to generate pseudo labels of the target-domain samples and construct a self-training set based on the confidence score. Second, we propose to use the spatio-temporal relationships among the samples in the non-self-training set to augment the core set selected by active learning. Finally, we combine the self-training set and the augmented core set to fine-tune the model. We demonstrate our method by comparing it with state-of-the-art methods on two IMU-based datasets and an EMG-based dataset. Our method presents similar HAR accuracies with the upper bound, i.e. fully supervised fine-tuning with less than 1\% labeled data of the target dataset and significantly improves data efficiency and time cost. Our work highlights the potential of implementing user-independent HAR methods into smart healthcare systems and BSN.
相關內容
Human Activity Recognition (HAR) has become a spotlight in recent scientific research because of its applications in various domains such as healthcare, athletic competitions, smart cities, and smart home. While researchers focus on the methodology of processing data, users wonder if the Artificial Intelligence (AI) methods used for HAR can be trusted. Trust depends mainly on the reliability or robustness of the system. To investigate the robustness of HAR systems, we analyzed several suitable current public datasets and selected WISDM for our investigation of Deep Learning approaches. While the published specification of WISDM matched our fundamental requirements (e.g., large, balanced, multi-hardware), several hidden issues were found in the course of our analysis. These issues reduce the performance and the overall trust of the classifier. By identifying the problems and repairing the dataset, the performance of the classifier was increased. This paper presents the methods by which other researchers may identify and correct similar problems in public datasets. By fixing the issues dataset veracity is improved, which increases the overall trust in the trained HAR system.
Few-shot text classification has recently been promoted by the meta-learning paradigm which aims to identify target classes with knowledge transferred from source classes with sets of small tasks named episodes. Despite their success, existing works building their meta-learner based on Prototypical Networks are unsatisfactory in learning discriminative text representations between similar classes, which may lead to contradictions during label prediction. In addition, the tasklevel and instance-level overfitting problems in few-shot text classification caused by a few training examples are not sufficiently tackled. In this work, we propose a contrastive learning framework named ContrastNet to tackle both discriminative representation and overfitting problems in few-shot text classification. ContrastNet learns to pull closer text representations belonging to the same class and push away text representations belonging to different classes, while simultaneously introducing unsupervised contrastive regularization at both task-level and instance-level to prevent overfitting. Experiments on 8 few-shot text classification datasets show that ContrastNet outperforms the current state-of-the-art models.
To investigate the well-observed racial disparities in computer vision systems that analyze images of humans, researchers have turned to skin tone as more objective annotation than race metadata for fairness performance evaluations. However, the current state of skin tone annotation procedures is highly varied. For instance, researchers use a range of untested scales and skin tone categories, have unclear annotation procedures, and provide inadequate analyses of uncertainty. In addition, little attention is paid to the positionality of the humans involved in the annotation process--both designers and annotators alike--and the historical and sociological context of skin tone in the United States. Our work is the first to investigate the skin tone annotation process as a sociotechnical project. We surveyed recent skin tone annotation procedures and conducted annotation experiments to examine how subjective understandings of skin tone are embedded in skin tone annotation procedures. Our systematic literature review revealed the uninterrogated association between skin tone and race and the limited effort to analyze annotator uncertainty in current procedures for skin tone annotation in computer vision evaluation. Our experiments demonstrated that design decisions in the annotation procedure such as the order in which the skin tone scale is presented or additional context in the image (i.e., presence of a face) significantly affected the resulting inter-annotator agreement and individual uncertainty of skin tone annotations. We call for greater reflexivity in the design, analysis, and documentation of procedures for evaluation using skin tone.
Research into the detection of human activities from wearable sensors is a highly active field, benefiting numerous applications, from ambulatory monitoring of healthcare patients via fitness coaching to streamlining manual work processes. We present an empirical study that compares 4 different commonly used annotation methods utilized in user studies that focus on in-the-wild data. These methods can be grouped in user-driven, in situ annotations - which are performed before or during the activity is recorded - and recall methods - where participants annotate their data in hindsight at the end of the day. Our study illustrates that different labeling methodologies directly impact the annotations' quality, as well as the capabilities of a deep learning classifier trained with the data respectively. We noticed that in situ methods produce less but more precise labels than recall methods. Furthermore, we combined an activity diary with a visualization tool that enables the participant to inspect and label their activity data. Due to the introduction of such a tool were able to decrease missing annotations and increase the annotation consistency, and therefore the F1-score of the deep learning model by up to 8% (ranging between 82.1 and 90.4% F1-score). Furthermore, we discuss the advantages and disadvantages of the methods compared in our study, the biases they may could introduce and the consequences of their usage on human activity recognition studies and as well as possible solutions.
Mining structured knowledge from tweets using named entity recognition (NER) can be beneficial for many downstream applications such as recommendation and intention under standing. With tweet posts tending to be multimodal, multimodal named entity recognition (MNER) has attracted more attention. In this paper, we propose a novel approach, which can dynamically align the image and text sequence and achieve the multi-level cross-modal learning to augment textual word representation for MNER improvement. To be specific, our framework can be split into three main stages: the first stage focuses on intra-modality representation learning to derive the implicit global and local knowledge of each modality, the second evaluates the relevance between the text and its accompanying image and integrates different grained visual information based on the relevance, the third enforces semantic refinement via iterative cross-modal interactions and co-attention. We conduct experiments on two open datasets, and the results and detailed analysis demonstrate the advantage of our model.
Individual differences of Electroencephalogram (EEG) could cause the domain shift which would significantly degrade the performance of cross-subject strategy. The domain adversarial neural networks (DANN), where the classification loss and domain loss jointly update the parameters of feature extractor, are adopted to deal with the domain shift. However, limited EEG data quantity and strong individual difference are challenges for the DANN with cumbersome feature extractor. In this work, we propose knowledge distillation (KD) based lightweight DANN to enhance cross-subject EEG-based emotion recognition. Specifically, the teacher model with strong context learning ability is utilized to learn complex temporal dynamics and spatial correlations of EEG, and robust lightweight student model is guided by the teacher model to learn more difficult domain-invariant features. In the feature-based KD framework, a transformer-based hierarchical temporalspatial learning model is served as the teacher model. The student model, which is composed of Bi-LSTM units, is a lightweight version of the teacher model. Hence, the student model could be supervised to mimic the robust feature representations of teacher model by leveraging complementary latent temporal features and spatial features. In the DANN-based cross-subject emotion recognition, we combine the obtained student model and a lightweight temporal-spatial feature interaction module as the feature extractor. And the feature aggregation is fed to the emotion classifier and domain classifier for domain-invariant feature learning. To verify the effectiveness of the proposed method, we conduct the subject-independent experiments on the public dataset DEAP with arousal and valence classification. The outstanding performance and t-SNE visualization of latent features verify the advantage and effectiveness of the proposed method.
Automatic Speech Recognition (ASR) systems exhibit the best performance on speech that is similar to that on which it was trained. As such, underrepresented varieties including regional dialects, minority-speakers, and low-resource languages, see much higher word error rates (WERs) than those varieties seen as 'prestigious', 'mainstream', or 'standard'. This can act as a barrier to incorporating ASR technology into the annotation process for large-scale linguistic research since the manual correction of the erroneous automated transcripts can be just as time and resource consuming as manual transcriptions. A deeper understanding of the behaviour of an ASR system is thus beneficial from a speech technology standpoint, in terms of improving ASR accuracy, and from an annotation standpoint, where knowing the likely errors made by an ASR system can aid in this manual correction. This work demonstrates a method of probing an ASR system to discover how it handles phonetic variation across a number of L2 Englishes. Specifically, how particular phonetic realisations which were rare or absent in the system's training data can lead to phoneme level misrecognitions and contribute to higher WERs. It is demonstrated that the behaviour of the ASR is systematic and consistent across speakers with similar spoken varieties (in this case the same L1) and phoneme substitution errors are typically in agreement with human annotators. By identifying problematic productions specific weaknesses can be addressed by sourcing such realisations for training and fine-tuning thus making the system more robust to pronunciation variation.
With the growing popularity of neural rendering, there has been an increasing number of neural implicit multi-view reconstruction methods. While many models have been enhanced in terms of positional encoding, sampling, rendering, and other aspects to improve the reconstruction quality, current methods do not fully leverage the information among neighboring pixels during the reconstruction process. To address this issue, we propose an enhanced model called BundleRecon. In the existing approaches, sampling is performed by a single ray that corresponds to a single pixel. In contrast, our model samples a patch of pixels using a bundle of rays, which incorporates information from neighboring pixels. Furthermore, we design bundle-based constraints to further improve the reconstruction quality. Experimental results demonstrate that BundleRecon is compatible with the existing neural implicit multi-view reconstruction methods and can improve their reconstruction quality.
Unsupervised domain adaptation has recently emerged as an effective paradigm for generalizing deep neural networks to new target domains. However, there is still enormous potential to be tapped to reach the fully supervised performance. In this paper, we present a novel active learning strategy to assist knowledge transfer in the target domain, dubbed active domain adaptation. We start from an observation that energy-based models exhibit free energy biases when training (source) and test (target) data come from different distributions. Inspired by this inherent mechanism, we empirically reveal that a simple yet efficient energy-based sampling strategy sheds light on selecting the most valuable target samples than existing approaches requiring particular architectures or computation of the distances. Our algorithm, Energy-based Active Domain Adaptation (EADA), queries groups of targe data that incorporate both domain characteristic and instance uncertainty into every selection round. Meanwhile, by aligning the free energy of target data compact around the source domain via a regularization term, domain gap can be implicitly diminished. Through extensive experiments, we show that EADA surpasses state-of-the-art methods on well-known challenging benchmarks with substantial improvements, making it a useful option in the open world. Code is available at //github.com/BIT-DA/EADA.
Deep learning-based semi-supervised learning (SSL) algorithms have led to promising results in medical images segmentation and can alleviate doctors' expensive annotations by leveraging unlabeled data. However, most of the existing SSL algorithms in literature tend to regularize the model training by perturbing networks and/or data. Observing that multi/dual-task learning attends to various levels of information which have inherent prediction perturbation, we ask the question in this work: can we explicitly build task-level regularization rather than implicitly constructing networks- and/or data-level perturbation-and-transformation for SSL? To answer this question, we propose a novel dual-task-consistency semi-supervised framework for the first time. Concretely, we use a dual-task deep network that jointly predicts a pixel-wise segmentation map and a geometry-aware level set representation of the target. The level set representation is converted to an approximated segmentation map through a differentiable task transform layer. Simultaneously, we introduce a dual-task consistency regularization between the level set-derived segmentation maps and directly predicted segmentation maps for both labeled and unlabeled data. Extensive experiments on two public datasets show that our method can largely improve the performance by incorporating the unlabeled data. Meanwhile, our framework outperforms the state-of-the-art semi-supervised medical image segmentation methods. Code is available at: //github.com/Luoxd1996/DTC
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191