Ye Tian

Cong Chen, Shouyang Dong, Ye Tian et al.

This paper focuses on Semi-Supervised Object Detection (SSOD). Knowledge Distillation (KD) has been widely used for semi-supervised image classification. However, adapting these methods for SSOD has the following obstacles. (1) The teacher model serves a dual role as a teacher and a student, such that the teacher predictions on unlabeled images may be very close to those of student, which limits the upper-bound of the student. (2) The class imbalance issue in SSOD hinders an efficient knowledge transfer from teacher to student. To address these problems, we propose a novel method Temporal Self-Ensembling Teacher (TSE-T) for SSOD. Differently from previous KD based methods, we devise a temporally evolved teacher model. First, our teacher model ensembles its temporal predictions for unlabeled images under stochastic perturbations. Second, our teacher model ensembles its temporal model weights with the student model weights by an exponential moving average (EMA) which allows the teacher gradually learn from the student. These self-ensembling strategies increase data and model diversity, thus improving teacher predictions on unlabeled images. Finally, we use focal loss to formulate consistency regularization term to handle the data imbalance problem, which is a more efficient manner to utilize the useful information from unlabeled images than a simple hard-thresholding method which solely preserves confident predictions. Evaluated on the widely used VOC and COCO benchmarks, the mAP of our method has achieved 80.73% and 40.52% on the VOC2007 test set and the COCO2014 minval5k set respectively, which outperforms a strong fully-supervised detector by 2.37% and 1.49%. Furthermore, our method sets the new state-of-the-art in SSOD on VOC2007 test set which outperforms the baseline SSOD method by 1.44%. The source code of this work is publicly available at http://github.com/syangdong/tse-t.

Chenyu Jiang, Ye Tian, Zhen Jia et al.

The Mixture-of-Expert (MoE) technique plays a crucial role in expanding the size of DNN model parameters. However, it faces the challenge of extended all-to-all communication latency during the training process. Existing methods attempt to mitigate this issue by overlapping all-to-all with expert computation. Yet, these methods frequently fall short of achieving sufficient overlap, consequently restricting the potential for performance enhancements. In our study, we extend the scope of this challenge by considering overlap at the broader training graph level. During the forward pass, we enable non-MoE computations to overlap with all-to-all through careful partitioning and pipelining. In the backward pass, we achieve overlap with all-to-all by scheduling gradient weight computations. We implement these techniques in Lancet, a system using compiler-based optimization to automatically enhance MoE model training. Our extensive evaluation reveals that Lancet significantly reduces the time devoted to non-overlapping communication, by as much as 77%. Moreover, it achieves a notable end-to-end speedup of up to 1.3 times when compared to the state-of-the-art solutions.

Ye Tian, Xiangxiang Chu, Hongpeng Wang

Most recent methods used for crowd counting are based on the convolutional neural network (CNN), which has a strong ability to extract local features. But CNN inherently fails in modeling the global context due to the limited receptive fields. However, the transformer can model the global context easily. In this paper, we propose a simple approach called CCTrans to simplify the design pipeline. Specifically, we utilize a pyramid vision transformer backbone to capture the global crowd information, a pyramid feature aggregation (PFA) model to combine low-level and high-level features, an efficient regression head with multi-scale dilated convolution (MDC) to predict density maps. Besides, we tailor the loss functions for our pipeline. Without bells and whistles, extensive experiments demonstrate that our method achieves new state-of-the-art results on several benchmarks both in weakly and fully-supervised crowd counting. Moreover, we currently rank No.1 on the leaderboard of NWPU-Crowd. Our code will be made available.

Jezabel R. Garcia, Federica Freddi, Feng-Ting Liao et al.

In meta-learning, the knowledge learned from previous tasks is transferred to new ones, but this transfer only works if tasks are related. Sharing information between unrelated tasks might hurt performance, and it is unclear how to transfer knowledge across tasks with a hierarchical structure. Our research extends a model agnostic meta-learning model, MAML, by exploiting hierarchical task relationships. Our algorithm, TreeMAML, adapts the model to each task with a few gradient steps, but the adaptation follows the hierarchical tree structure: in each step, gradients are pooled across tasks clusters, and subsequent steps follow down the tree. We also implement a clustering algorithm that generates the tasks tree without previous knowledge of the task structure, allowing us to make use of implicit relationships between the tasks. We show that the new algorithm, which we term TreeMAML, performs better than MAML when the task structure is hierarchical for synthetic experiments. To study the performance of the method in real-world data, we apply this method to Natural Language Understanding, we use our algorithm to finetune Language Models taking advantage of the language phylogenetic tree. We show that TreeMAML improves the state of the art results for cross-lingual Natural Language Inference. This result is useful, since most languages in the world are under-resourced and the improvement on cross-lingual transfer allows the internationalization of NLP models. This results open the window to use this algorithm in other real-world hierarchical datasets.

Cheng Li, Ye Tian

Supervised relation extraction methods based on deep neural network play an important role in the recent information extraction field. However, at present, their performance still fails to reach a good level due to the existence of complicated relations. On the other hand, recently proposed pre-trained language models (PLMs) have achieved great success in multiple tasks of natural language processing through fine-tuning when combined with the model of downstream tasks. However, original standard tasks of PLM do not include the relation extraction task yet. We believe that PLMs can also be used to solve the relation extraction problem, but it is necessary to establish a specially designed downstream task model or even loss function for dealing with complicated relations. In this paper, a new network architecture with a special loss function is designed to serve as a downstream model of PLMs for supervised relation extraction. Experiments have shown that our method significantly exceeded the current optimal baseline models across multiple public datasets of relation extraction.

Wei Wang, Ye Tian, Jiquan Ngiam et al.

Most data selection research in machine translation focuses on improving a single domain. We perform data selection for multiple domains at once. This is achieved by carefully introducing instance-level domain-relevance features and automatically constructing a training curriculum to gradually concentrate on multi-domain relevant and noise-reduced data batches. Both the choice of features and the use of curriculum are crucial for balancing and improving all domains, including out-of-domain. In large-scale experiments, the multi-domain curriculum simultaneously reaches or outperforms the individual performance and brings solid gains over no-curriculum training.

Lu Haonan, Seth H. Huang, Tian Ye et al.

In this work, we present graph star net (GraphStar), a novel and unified graph neural net architecture which utilizes message-passing relay and attention mechanism for multiple prediction tasks - node classification, graph classification and link prediction. GraphStar addresses many earlier challenges facing graph neural nets and achieves non-local representation without increasing the model depth or bearing heavy computational costs. We also propose a new method to tackle topic-specific sentiment analysis based on node classification and text classification as graph classification. Our work shows that 'star nodes' can learn effective graph-data representation and improve on current methods for the three tasks. Specifically, for graph classification and link prediction, GraphStar outperforms the current state-of-the-art models by 2-5% on several key benchmarks.

Ye Tian, Li Yang, Wei Wang et al.

Goal: Squamous cell carcinoma of cervix is one of the most prevalent cancer worldwide in females. Traditionally, the most indispensable diagnosis of cervix squamous carcinoma is histopathological assessment which is achieved under microscope by pathologist. However, human evaluation of pathology slide is highly depending on the experience of pathologist, thus big inter- and intra-observer variability exists. Digital pathology, in combination with deep learning provides an opportunity to improve the objectivity and efficiency of histopathologic slide analysis. Methods: In this study, we obtained 800 haematoxylin and eosin stained slides from 300 patients suffered from cervix squamous carcinoma. Based on information from morphological heterogeneity in the tumor and its adjacent area, we established deep learning models using popular convolution neural network architectures (inception-v3, InceptionResnet-v2 and Resnet50). Then random forest was introduced to feature extractions and slide-based classification. Results: The overall performance of our proposed models on slide-based tumor discrimination were outstanding with an AUC scores > 0.94. While, location identifications of lesions in whole slide images were mediocre (FROC scores > 0.52) duo to the extreme complexity of tumor tissues. Conclusion: For the first time, our analysis workflow highlighted a quantitative visual-based slide analysis of cervix squamous carcinoma. Significance: This study demonstrates a pathway to assist pathologist and accelerate the diagnosis of patients by utilizing new computational approaches.

Ye Tian, Weiping Zhang

In this paper, we propose an effective THresholding method based on ORder Statistic, called THORS, to convert an arbitrary scoring-type classifier, which can induce a continuous cumulative distribution function of the score, into a cost-sensitive one. The procedure, uses order statistic to find an optimal threshold for classification, requiring almost no knowledge of classifiers itself. Unlike common data-driven methods, we analytically show that THORS has theoretical guaranteed performance, theoretical bounds for the costs and lower time complexity. Coupled with empirical results on several real-world data sets, we argue that THORS is the preferred cost-sensitive technique.

Yunhan Zhao, Ye Tian, Charless Fowlkes et al.

Training deep networks that generalize to a wide range of variations in test data is essential to building accurate and robust image classifiers. One standard strategy is to apply data augmentation to synthetically enlarge the training set. However, data augmentation is essentially a brute-force method which generates uniform samples from some pre-defined set of transformations. In this paper, we propose a principled approach to train networks with significantly improved resistance to large variations between training and testing data. This is achieved by embedding a learnable transformation module into the introspective network, which is a convolutional neural network (CNN) classifier empowered with generative capabilities. Our approach alternates between synthesizing pseudo-negative samples and transformed positive examples based on the current model, and optimizing model predictions on these synthesized samples. Experimental results verify that our approach significantly improves the ability of deep networks to resist large variations between training and testing data and achieves classification accuracy improvements on several benchmark datasets, including MNIST, affNIST, SVHN, CIFAR-10 and miniImageNet.