Chinese Spelling Correction (CSC) is the task of detecting and correcting misspelled characters in Chinese texts.As an important step for various downstream tasks, CSC confronts two challenges: 1) Character-level errors consist not only of spelling errors but also of missing and redundant ones that cause variable length between input and output texts, for which most CSC methods could not handle well because of the consistence length of texts required by their inherent detection-correction framework.Consequently, the two errors are considered outside the scope and left to future work, despite the fact that they are widely found and bound to CSC task in Chinese industrial scenario, such as Automatic Speech Recognition (ASR) and Optical Character Recognition (OCR).2) Most existing CSC methods focus on either detector or corrector and train different models for each one, respectively, leading to insufficiency of parameters sharing.To address these issues, we propose a novel model UMR-Spell to learn detection and correction parts together at the same time from a multi-task learning perspective by using a detection transmission self-attention matrix, and flexibly deal with both missing, redundant, and spelling errors through re-tagging rules.Furthermore, we build a new dataset ECMR-2023 containing five kinds of character-level errors to enrich the CSC task closer to real-world applications.Experiments on both SIGHAN benchmarks and ECMR-2023 demonstrate the significant effectiveness of UMRSpell over previous representative baselines.

Tabular data synthesis is an emerging approach to circumvent strict regulations on data privacy while discovering knowledge through big data. Although state-of-the-art AI-based tabular data synthesizers, e.g., table-GAN, CTGAN, TVAE, and CTAB-GAN, are effective at generating synthetic tabular data, their training is sensitive to column permutations of input data. In this paper, we first c onduct a n e xtensive e mpirical s tudy to disclose such a property of permutation invariance and an in-depth analysis of the existing synthesizers. We show that changing the input column order worsens the statistical difference between real and synthetic data by up to 38.67% due to the encoding of tabular data and the network architectures. To fully unleash the potential of big synthetic tabular data, we propose two solutions: (i) AE-GAN, a synthesizer that uses an autoencoder network to represent the tabular data and GAN networks to synthesize the latent representation, and (ii) a feature sorting algorithm to find t he s uitable c olumn o rder o f i nput d ata f or CNN-based synthesizers. We evaluate the proposed solutions on five datasets in terms of the sensitivity to the column permutation, the quality of synthetic data, and the utility in downstream analyses. Our results show that we enhance the property of permutation-invariance when training synthesizers and further improve the quality and utility of synthetic data, up to 22%, compared to the existing synthesizers.

Continual relation extraction is an important task that focuses on extracting new facts incrementally from unstructured text. Given the sequential arrival order of the relations, this task is prone to two serious challenges, namely catastrophic forgetting and order-sensitivity. We propose a novel curriculum-meta learning method to tackle the above two challenges in continual relation extraction. We combine meta learning and curriculum learning to quickly adapt model parameters to a new task and to reduce interference of previously seen tasks on the current task. We design a novel relation representation learning method through the distribution of domain and range types of relations. Such representations are utilized to quantify the difficulty of tasks for the construction of curricula. Moreover, we also present novel difficulty-based metrics to quantitatively measure the extent of order-sensitivity of a given model, suggesting new ways to evaluate model robustness. Our comprehensive experiments on three benchmark datasets show that our proposed method outperforms the state-of-the-art techniques. The code is available at the anonymous GitHub repository: https://github.com/wutong8023/AAAI_CML.

The classification on imbalanced data sets is a great challenge in machine learning. In this paper, a geometric structural ensemble (GSE) learning framework is proposed to address the issue. It is known that the traditional ensemble methods train and combine a series of basic classifiers according to various weights, which might lack the geometric meaning. Oppositely, the GSE partitions and eliminates redundant majority samples by generating hyper-sphere through the Euclidean metric and learns basic classifiers to enclose the minority samples, which achieves higher efficiency in the training process and seems easier to understand. In detail, the current weak classifier builds boundaries between the majority and the minority samples and removes the former. Then, the remaining samples are used to train the next. When the training process is done, all of the majority samples could be cleaned and the combination of all basic classifiers is obtained. To further improve the generalization, two relaxation techniques are proposed. Theoretically, the computational complexity of GSE could approach O(nd log(nmin) log(nmaj)). The comprehensive experiments validate both the effectiveness and efficiency of GSE.

In this article, we propose a novel entropy and confidence-based undersampling boosting (ECUBoost) framework to solve imbalanced problems. The boosting-based ensemble is combined with a new undersampling method to improve the generalization performance. To avoid losing informative samples during the data preprocessing of the boosting-based ensemble, both confidence and entropy are used in ECUBoost as benchmarks to ensure the validity and structural distribution of the majority samples during the undersampling. Furthermore, different from other iterative dynamic resampling methods, ECUBoost based on confidence can be applied to algorithms without iterations such as decision trees. Meanwhile, random forests are used as base classifiers in ECUBoost. Furthermore, experimental results on both artificial data sets and KEEL data sets prove the effectiveness of the proposed method.

Existing learning models for classification of imbalanced data sets can be grouped as either boundary-based or nonboundary-based depending on whether a decision hyperplane is used in the learning process. The focus of this paper is a new approach that leverages the advantage of both approaches. Specifically, our new model partitions the input space into three parts by creating two additional boundaries in the training process, and then makes the final decision based on a heuristic measurement between the test sample and a subset of selected training samples. Since the original hyperplane used by the underlying original classifier will be eliminated, the proposed model is named the boundary-eliminated (BE) model. Additionally, the pseudoinverse linear discriminant (PILD) is adopted for the BE model so as to obtain a novel classifier abbreviated as BEPILD. Experiments validate both the effectiveness and the efficiency of BEPILD, compared with 13 state-of-the-art classification methods, based on 31 imbalanced and 7 standard data sets.

From the perspective of economic and efficient benefits, modern data center oversubscribes power supplies to deploy as many servers as possible. The oversubscription is based on the varied loads among servers to modulate power demand. Nevertheless, power oversubscription has potential threats to system availability, the data center may collapse as a result of overloading. Current solutions to the oversubscription usually focus on managing the datacenter workload to avoid the peak power demand time in the data center. However, none of the current research considers the influence of the failure of the power or utility components, where the component failure may affect the effectiveness of these strategies. Meanwhile, none of these current research can answer the question that how many servers should be deployed in the data centers under an availability constraint. In this paper, we propose quantitative availability analysis of hierarchical datacenter under power oversubscription. To this end, we use Markov chain and Stochastic Reward Net (SRN) to model the failure and repair processes of data center components. The servers at the bottom level are distributed in two pools: main pool and backup pool, where running servers are in main pool and turned-off servers in backup pool. Migration from backup pool to main pool is conducted once any running server fails. SRNs are implemented to model these two pools, and Markov chain is used to model the components in the upper level. The evaluation is based on the real-life Google and Wikipedia traces. The result shows the relationship between oversubscription and data center availability, which can guide the data center operators to choose the appropriate oversubscription ratio under the availability constraint.