Learning Deep Landmarks for Imbalanced Classification

Emotion-Semantic-Aware Dual Contrastive Learning for Epistemic Emotion Identification of Learner-Generated Reviews in MOOCs

Identifying the epistemic emotions of learner-generated reviews in massive open online courses (MOOCs) can help instructors provide adaptive guidance and interventions for learners. The epistemic emotion identification task is a fine-grained identification task that contains multiple categories of emotions arising during the learning process. Previous studies only consider emotional or semantic information within the review texts alone, which leads to insufficient feature representation. In addition, some categories of epistemic emotions are ambiguously distributed in feature space, making them hard to be distinguished. In this article, we present an emotion-semantic-aware dual contrastive learning (ES-DCL) approach to tackle these issues. In order to learn sufficient feature representation, implicit semantic features and human-interpretable emotional features are, respectively, extracted from two different views to form complementary emotional-semantic features. On this basis, by leveraging the experience of domain experts and the input emotional-semantic features, two types of contrastive losses (label contrastive loss and feature contrastive loss) are formulated. They are designed to train the discriminative distribution of emotional-semantic features in the sample space and to solve the anisotropy problem between different categories of epistemic emotions. The proposed ES-DCL is compared with 11 other baseline models on four different disciplinary MOOCs review datasets. Extensive experimental results show that our approach improves the performance of epistemic emotion identification, and significantly outperforms state-of-the-art deep learning-based methods in learning more discriminative sentence representations.

A Comprehensive Framework for Long-Tailed Learning via Pretraining and Normalization

Data in the visual world often present long-tailed distributions. However, learning high-quality representations and classifiers for imbalanced data is still challenging for data-driven deep learning models. In this work, we aim at improving the feature extractor and classifier for long-tailed recognition via contrastive pretraining and feature normalization, respectively. First, we carefully study the influence of contrastive pretraining under different conditions, showing that current self-supervised pretraining for long-tailed learning is still suboptimal in both performance and speed. We thus propose a new balanced contrastive loss and a fast contrastive initialization scheme to improve previous long-tailed pretraining. Second, based on the motivative analysis on the normalization for classifier, we propose a novel generalized normalization classifier that consists of generalized normalization and grouped learnable scaling. It outperforms traditional inner product classifier as well as cosine classifier. Both the two components proposed can improve recognition ability on tail classes without the expense of head classes. We finally build a unified framework that achieves competitive performance compared with state of the arts on several long-tailed recognition benchmarks and maintains high efficiency.

RDDL: A systematic ensemble pipeline tool that streamlines balancing training schemes to reduce the effects of data imbalance in rare-disease-related deep-learning applications

Identifying lowly prevalent diseases, or rare diseases, in their early stages is key to disease treatment in the medical field. Deep learning techniques now provide promising tools for this purpose. Nevertheless, the low prevalence of rare diseases entangles the proper application of deep networks for disease identification due to the severe class-imbalance issue. In the past decades, some balancing methods have been studied to handle the data-imbalance issue. The bad news is that it is verified that none of these methods guarantees superior performance to others. This performance variation causes the need to formulate a systematic pipeline with a comprehensive software tool for enhancing deep-learning applications in rare disease identification. We reviewed the existing balancing schemes and summarized a systematic deep ensemble pipeline with a constructed tool called RDDL for handling the data imbalance issue. Through two real case studies, we showed that rare disease identification could be boosted with this systematic RDDL pipeline tool by lessening the data imbalance problem during model training. The RDDL pipeline tool is available at https://github.com/cobisLab/RDDL/.

DeepSMOTE: Fusing Deep Learning and SMOTE for Imbalanced Data

Despite over two decades of progress, imbalanced data is still considered a significant challenge for contemporary machine learning models. Modern advances in deep learning have further magnified the importance of the imbalanced data problem, especially when learning from images. Therefore, there is a need for an oversampling method that is specifically tailored to deep learning models, can work on raw images while preserving their properties, and is capable of generating high-quality, artificial images that can enhance minority classes and balance the training set. We propose Deep synthetic minority oversampling technique (SMOTE), a novel oversampling algorithm for deep learning models that leverages the properties of the successful SMOTE algorithm. It is simple, yet effective in its design. It consists of three major components: 1) an encoder/decoder framework; 2) SMOTE-based oversampling; and 3) a dedicated loss function that is enhanced with a penalty term. An important advantage of DeepSMOTE over generative adversarial network (GAN)-based oversampling is that DeepSMOTE does not require a discriminator, and it generates high-quality artificial images that are both information-rich and suitable for visual inspection. DeepSMOTE code is publicly available at https://github.com/dd1github/DeepSMOTE.

Imbalanced prediction of emergency department admission using natural language processing and deep neural network

The emergency department (ED) plays a very significant role in the hospital. Owing to the rising number of ED visits, medical service points, and ED market, overcrowding of EDs has become serious worldwide. Overcrowding has long been recognized as a vital issue that increases the risk to patients and negative emotions of medical personnel and impacts hospital cost management. For the past years, many researchers have been applying artificial intelligence to reduce crowding situations in the ED. Nevertheless, the datasets in ED hospital admission are naturally inherent with the high-class imbalance in the real world. Previous studies have not considered the imbalance of the datasets, particularly addressing the imbalance. This study purposes to develop a natural language processing model of a deep neural network with an attention mechanism to solve the imbalanced problem in ED admission. The proposed framework is used for predicting hospital admission so that the hospitals can arrange beds early and solve the problem of congestion in the ED. Furthermore, the study compares a variety of methods and obtains the best composition that has the best performance for forecasting hospitalization in ED. The study used the data from a specific hospital in Taiwan as an empirical study. The experimental result demonstrates that almost all imbalanced methods can improve the model's performance. In addition, the natural language processing model of Bi-directional Long Short-Term Memory with attention mechanism has the best results in all-natural language processing methods.

Human-in-the-Loop Low-Shot Learning

We consider a human-in-the-loop scenario in the context of low-shot learning. Our approach was inspired by the fact that the viability of samples in novel categories cannot be sufficiently reflected by those limited observations. Some heterogeneous samples that are quite different from existing labeled novel data can inevitably emerge in the testing phase. To this end, we consider augmenting an uncertainty assessment module into low-shot learning system to account into the disturbance of those out-of-distribution (OOD) samples. Once detected, these OOD samples are passed to human beings for active labeling. Due to the discrete nature of this uncertainty assessment process, the whole Human-In-the-Loop Low-shot (HILL) learning framework is not end-to-end trainable. We hence revisited the learning system from the aspect of reinforcement learning and introduced the REINFORCE algorithm to optimize model parameters via policy gradient. The whole system gains noticeable improvements over existing low-shot learning approaches.

Spatio-temporal graph convolutional network for diagnosis and treatment response prediction of major depressive disorder from functional connectivity

The pathophysiology of major depressive disorder (MDD) has been explored to be highly associated with the dysfunctional integration of brain networks. It is therefore imperative to explore neuroimaging biomarkers to aid diagnosis and treatment. In this study, we developed a spatiotemporal graph convolutional network (STGCN) framework to learn discriminative features from functional connectivity for automatic diagnosis and treatment response prediction of MDD. Briefly, dynamic functional networks were first obtained from the resting-state fMRI with the sliding temporal window method. Secondly, a novel STGCN approach was proposed by introducing the modules of spatial graph attention convolution (SGAC) and temporal fusion. A novel SGAC was proposed to improve the feature learning ability and special anatomy prior guided pooling was developed to enable the feature dimension reduction. A temporal fusion module was proposed to capture the dynamic features of functional connectivity between adjacent sliding windows. Finally, the STGCN proposed approach was utilized to the tasks of diagnosis and antidepressant treatment response prediction for MDD. Performances of the framework were comprehensively examined with large cohorts of clinical data, which demonstrated its effectiveness in classifying MDD patients and predicting the treatment response. The sound performance suggests the potential of the STGCN for the clinical use in diagnosis and treatment prediction.