We propose to jointly detect and classify emergency events using a multi-class text classifier, which is a typical deep learning architecture with transformer modules and particularly employs Bidirectional Encoder Rep...
详细信息
ISBN:
(纸本)9781665440899
We propose to jointly detect and classify emergency events using a multi-class text classifier, which is a typical deep learning architecture with transformer modules and particularly employs Bidirectional Encoder Representations from Transformers(BERT). Deep learning requires a large number of labeled data to work. Meanwhile, deep learning often implements the semi-supervised learning(SSL) method, which is able to use massive unlabeled data to improve performance of supervised deep learning. As an effective SSL variant, unsupervised data augmentation (UDA) focuses on dataaugmentation techniques to improve the performance of deep learning. We present an enhanced version of UDA(EUDA) by mixing more dataaugmentation strategies and using a problem related prefilter. Our EUDA targets at emergency event detection and classification. Considering that emergency events always have time and location elements, text can be filtered based on this semantic feature. We propose to use semantic feature aided enhanced unsupervised data augmentation to solve the concerned problem. Empirical studies on the dataset prepared for the task validates that the proposed EUDA can achieve significantly better performance than supervised learning with a limited size of labeled data. Experiments are also carried out on a text classification task, which confirms that EUDA improves performance for BERT neural network.
Osteoarthritis (OA) is a chronic degenerative disorder of joints and is the most common reason leading to total knee joint replacement (TKR). In this paper, we implemented a semi-supervised learning approach based on ...
详细信息
ISBN:
(纸本)9781510633964
Osteoarthritis (OA) is a chronic degenerative disorder of joints and is the most common reason leading to total knee joint replacement (TKR). In this paper, we implemented a semi-supervised learning approach based on unsupervised data augmentation (UDA) along with valid perturbations for radiographs to enhance the performance of supervised TKR outcome prediction model. Our results suggest that the use of semi-supervised approach provides superior results compared to the supervised approach (AUC of 0.79 +/- 0.04 vs 0.74 +/- 0.04).
We propose to jointly detect and classify emergency events using a multi-class text classifier,which is a typical deep learning architecture with transformer modules and particularly employs Bidirectional Encoder Repr...
详细信息
We propose to jointly detect and classify emergency events using a multi-class text classifier,which is a typical deep learning architecture with transformer modules and particularly employs Bidirectional Encoder Representations from Transformers(BERT).Deep learning requires a large number of labeled data to ***,deep learning often implements the semi-supervised learning(SSL) method,which is able to use massive unlabeled data to improve performance of supervised deep *** an effective SSL variant,unsupervised data augmentation(UDA) focuses on dataaugmentation techniques to improve the performance of deep *** present an enhanced version of UDA(EUDA) by mixing more dataaugmentation strategies and using a problem related *** EUDA targets at emergency event detection and *** that emergency events always have time and location elements,text can be filtered based on this semantic *** propose to use semantic feature aided enhanced unsupervised data augmentation to solve the concerned *** studies on the dataset prepared for the task validates that the proposed EUDA can achieve significantly better performance than supervised learning with a limited size of labeled *** are also carried out on a text classification task,which confirms that EUDA improves performance for BERT neural network.
Infrastructure inspections generally cannot provide sufficient data for model training to automate its procedure of damage detection. Therefore, this study proposed a two-stepped automated concrete crack segmentation ...
详细信息
Infrastructure inspections generally cannot provide sufficient data for model training to automate its procedure of damage detection. Therefore, this study proposed a two-stepped automated concrete crack segmentation method under small datasets. Firstly, the generative model, namely the Progressive Growing Denoising Diffusion Probabilistic Model (PG-DDPM), trained by real crack images, was utilised to create pseudo-crack images for the unsupervised data augmentation to address the deficiency of datasets. Secondly, A crack segmentation model based on DeepLabv3+ network was trained with datasets mixed with real and generated crack images to verify the effectiveness of the PG-DDPM-augmented datasets. The experimental results demonstrate that when the original crack dataset is composed of 400 real images, adding 400 generated crack images improves the mPA of the DeepLabv3+ network by approximately 1.4 %. On the other hand, adding 1200 generated crack images to the dataset composed of 1200 real images marginally improves the mPA by approximately 0.6 %.
The current trend in automatic speech recognition is to leverage large amounts of labeled data to train supervised neural network models. Unfortunately, obtaining data for a wide range of domains to train robust model...
详细信息
ISBN:
(纸本)9781510872219
The current trend in automatic speech recognition is to leverage large amounts of labeled data to train supervised neural network models. Unfortunately, obtaining data for a wide range of domains to train robust models can be costly. However, it is relatively inexpensive to collect large amounts of unlabeled data from domains that we want the models to generalize to. In this paper, we propose a novel unsupervised adaptation method that learns to synthesize labeled data for the target domain from unlabeled in-domain data and labeled out-of-domain data. We first learn without supervision an interpretable latent representation of speech that encodes linguistic and nuisance factors (e.g., speaker and channel) using different latent variables. To transform a labeled out-of-domain utterance without altering its transcript, we transform the latent nuisance variables while maintaining the linguistic variables. To demonstrate our approach, we focus on a channel mismatch setting, where the domain of interest is distant conversational speech, and labels are only available for close-talking speech. Our proposed method is evaluated on the AMI dataset, outperforming all baselines and bridging the gap between unadapted and in-domain models by over 77% without using any parallel data.
暂无评论