Research papers and code for "Hui Jiang":
In this paper, we present some theoretical work to explain why simple gradient descent methods are so successful in solving non-convex optimization problems in learning large-scale neural networks (NN). After introducing a mathematical tool called canonical space, we have proved that the objective functions in learning NNs are convex in the canonical model space. We further elucidate that the gradients between the original NN model space and the canonical space are related by a pointwise linear transformation, which is represented by the so-called disparity matrix. Furthermore, we have proved that gradient descent methods surely converge to a global minimum of zero loss provided that the disparity matrices maintain full rank. If this full-rank condition holds, the learning of NNs behaves in the same way as normal convex optimization. At last, we have shown that the chance to have singular disparity matrices is extremely slim in large NNs. In particular, when over-parameterized NNs are randomly initialized, the gradient decent algorithms converge to a global minimum of zero loss in probability.

* 10 pages
Click to Read Paper and Get Code
In this work, we introduce the concept of bandlimiting into the theory of machine learning because all physical processes are bandlimited by nature, including real-world machine learning tasks. After the bandlimiting constraint is taken into account, our theoretical analysis has shown that all practical machine learning tasks are asymptotically solvable in a perfect sense. Furthermore, the key towards this solvability almost solely relies on two factors: i) a sufficiently large amount of training samples beyond a threshold determined by a difficulty measurement of the underlying task; ii) a sufficiently complex model that is properly bandlimited. Moreover, for some special cases, we have derived new error bounds for perfect learning, which can quantify the difficulty of learning. These case-specific bounds are much tighter than the uniform bounds in conventional learning theory. Our results have provided a new perspective to explain the recent successes of large-scale supervised learning using complex models like neural networks.

* 9 pages
Click to Read Paper and Get Code
Content-based news recommendation systems need to recommend news articles based on the topics and content of articles without using user specific information. Many news articles describe the occurrence of specific events and named entities including people, places or objects. In this paper, we propose a graph traversal algorithm as well as a novel weighting scheme for cold-start content based news recommendation utilizing these named entities. Seeking to create a higher degree of user-specific relevance, our algorithm computes the shortest distance between named entities, across news articles, over a large knowledge graph. Moreover, we have created a new human annotated data set for evaluating content based news recommendation systems. Experimental results show our method is suitable to tackle the hard cold-start problem and it produces stronger Pearson correlation to human similarity scores than other cold-start methods. Our method is also complementary and a combination with the conventional cold-start recommendation methods may yield significant performance gains. The dataset, CNRec, is available at: https://github.com/kevinj22/CNRec

Click to Read Paper and Get Code
To apply general knowledge to machine reading comprehension (MRC), we propose an innovative MRC approach, which consists of a WordNet-based data enrichment method and an MRC model named as Knowledge Aided Reader (KAR). The data enrichment method uses the semantic relations of WordNet to extract semantic level inter-word connections from each passage-question pair in the MRC dataset, and allows us to control the amount of the extraction results by setting a hyper-parameter. KAR uses the extraction results of the data enrichment method as explicit knowledge to assist the prediction of answer spans. According to the experimental results, the single model of KAR achieves an Exact Match (EM) of $72.4$ and an F1 Score of $81.1$ on the development set of SQuAD, and more importantly, by applying different settings in the data enrichment method to change the amount of the extraction results, there is a $2\%$ variation in the resulting performance of KAR, which implies that the explicit knowledge provided by the data enrichment method plays an effective role in the training of KAR.

Click to Read Paper and Get Code
As a generative model for building end-to-end dialogue systems, Hierarchical Recurrent Encoder-Decoder (HRED) consists of three layers of Gated Recurrent Unit (GRU), which from bottom to top are separately used as the word-level encoder, the sentence-level encoder, and the decoder. Despite performing well on dialogue corpora, HRED is computationally expensive to train due to its complexity. To improve the training efficiency of HRED, we propose a new model, which is named as Simplified HRED (SHRED), by making each layer of HRED except the top one simpler than its upper layer. On the one hand, we propose Scalar Gated Unit (SGU), which is a simplified variant of GRU, and use it as the sentence-level encoder. On the other hand, we use Fixed-size Ordinally-Forgetting Encoding (FOFE), which has no trainable parameter at all, as the word-level encoder. The experimental results show that compared with HRED under the same word embedding size and the same hidden state size for each layer, SHRED reduces the number of trainable parameters by 25\%--35\%, and the training time by more than 50\%, but still achieves slightly better performance.

Click to Read Paper and Get Code
In the past few years, Generative Adversarial Network (GAN) became a prevalent research topic. By defining two convolutional neural networks (G-Network and D-Network) and introducing an adversarial procedure between them during the training process, GAN has ability to generate good quality images that look like natural images from a random vector. Besides image generation, GAN may have potential to deal with wide range of real world problems. In this paper, we follow the basic idea of GAN and propose a novel model for image saliency detection, which is called Supervised Adversarial Networks (SAN). Specifically, SAN also trains two models simultaneously: the G-Network takes natural images as inputs and generates corresponding saliency maps (synthetic saliency maps), and the D-Network is trained to determine whether one sample is a synthetic saliency map or ground-truth saliency map. However, different from GAN, the proposed method uses fully supervised learning to learn both G-Network and D-Network by applying class labels of the training set. Moreover, a novel kind of layer call conv-comparison layer is introduced into the D-Network to further improve the saliency performance by forcing the high-level feature of synthetic saliency maps and ground-truthes as similar as possible. Experimental results on Pascal VOC 2012 database show that the SAN model can generate high quality saliency maps for many complicate natural images.

Click to Read Paper and Get Code
In this paper, we study a novel approach for named entity recognition (NER) and mention detection in natural language processing. Instead of treating NER as a sequence labelling problem, we propose a new local detection approach, which rely on the recent fixed-size ordinally forgetting encoding (FOFE) method to fully encode each sentence fragment and its left/right contexts into a fixed-size representation. Afterwards, a simple feedforward neural network is used to reject or predict entity label for each individual fragment. The proposed method has been evaluated in several popular NER and mention detection tasks, including the CoNLL 2003 NER task and TAC-KBP2015 and TAC-KBP2016 Tri-lingual Entity Discovery and Linking (EDL) tasks. Our methods have yielded pretty strong performance in all of these examined tasks. This local detection approach has shown many advantages over the traditional sequence labelling methods.

Click to Read Paper and Get Code
Convolutional neural networks (CNNs) have yielded the excellent performance in a variety of computer vision tasks, where CNNs typically adopt a similar structure consisting of convolution layers, pooling layers and fully connected layers. In this paper, we propose to apply a novel method, namely Hybrid Orthogonal Projection and Estimation (HOPE), to CNNs in order to introduce orthogonality into the CNN structure. The HOPE model can be viewed as a hybrid model to combine feature extraction using orthogonal linear projection with mixture models. It is an effective model to extract useful information from the original high-dimension feature vectors and meanwhile filter out irrelevant noises. In this work, we present three different ways to apply the HOPE models to CNNs, i.e., {\em HOPE-Input}, {\em single-HOPE-Block} and {\em multi-HOPE-Blocks}. For {\em HOPE-Input} CNNs, a HOPE layer is directly used right after the input to de-correlate high-dimension input feature vectors. Alternatively, in {\em single-HOPE-Block} and {\em multi-HOPE-Blocks} CNNs, we consider to use HOPE layers to replace one or more blocks in the CNNs, where one block may include several convolutional layers and one pooling layer. The experimental results on both Cifar-10 and Cifar-100 data sets have shown that the orthogonal constraints imposed by the HOPE layers can significantly improve the performance of CNNs in these image classification tasks (we have achieved one of the best performance when image augmentation has not been applied, and top 5 performance with image augmentation).

* 7 Pages, 5 figures, submitted to AAAI 2017
Click to Read Paper and Get Code
In this paper, we study novel neural network structures to better model long term dependency in sequential data. We propose to use more memory units to keep track of more preceding states in recurrent neural networks (RNNs), which are all recurrently fed to the hidden layers as feedback through different weighted paths. By extending the popular recurrent structure in RNNs, we provide the models with better short-term memory mechanism to learn long term dependency in sequences. Analogous to digital filters in signal processing, we call these structures as higher order RNNs (HORNNs). Similar to RNNs, HORNNs can also be learned using the back-propagation through time method. HORNNs are generally applicable to a variety of sequence modelling tasks. In this work, we have examined HORNNs for the language modeling task using two popular data sets, namely the Penn Treebank (PTB) and English text8 data sets. Experimental results have shown that the proposed HORNNs yield the state-of-the-art performance on both data sets, significantly outperforming the regular RNNs as well as the popular LSTMs.

* 9 pages
Click to Read Paper and Get Code
In this paper, we propose a fast deep learning method for object saliency detection using convolutional neural networks. In our approach, we use a gradient descent method to iteratively modify the input images based on the pixel-wise gradients to reduce a pre-defined cost function, which is defined to measure the class-specific objectness and clamp the class-irrelevant outputs to maintain image background. The pixel-wise gradients can be efficiently computed using the back-propagation algorithm. We further apply SLIC superpixels and LAB color based low level saliency features to smooth and refine the gradients. Our methods are quite computationally efficient, much faster than other deep learning based saliency methods. Experimental results on two benchmark tasks, namely Pascal VOC 2012 and MSRA10k, have shown that our proposed methods can generate high-quality salience maps, at least comparable with many slow and complicated deep learning methods. Comparing with the pure low-level methods, our approach excels in handling many difficult images, which contain complex background, highly-variable salient objects, multiple objects, and/or very small salient objects.

* arXiv admin note: substantial text overlap with arXiv:1505.01173
Click to Read Paper and Get Code
Financial news contains useful information on public companies and the market. In this paper we apply the popular word embedding methods and deep neural networks to leverage financial news to predict stock price movements in the market. Experimental results have shown that our proposed methods are simple but very effective, which can significantly improve the stock prediction accuracy on a standard financial database over the baseline system using only the historical price information.

* 5 pages, 2 figures, technical report
Click to Read Paper and Get Code
In this paper, we propose a novel model for high-dimensional data, called the Hybrid Orthogonal Projection and Estimation (HOPE) model, which combines a linear orthogonal projection and a finite mixture model under a unified generative modeling framework. The HOPE model itself can be learned unsupervised from unlabelled data based on the maximum likelihood estimation as well as discriminatively from labelled data. More interestingly, we have shown the proposed HOPE models are closely related to neural networks (NNs) in a sense that each hidden layer can be reformulated as a HOPE model. As a result, the HOPE framework can be used as a novel tool to probe why and how NNs work, more importantly, to learn NNs in either supervised or unsupervised ways. In this work, we have investigated the HOPE framework to learn NNs for several standard tasks, including image recognition on MNIST and speech recognition on TIMIT. Experimental results have shown that the HOPE framework yields significant performance gains over the current state-of-the-art methods in various types of NN learning problems, including unsupervised feature learning, supervised or semi-supervised learning.

* Journal of Machine Learning Research (JMLR), 17(37):1-33, 2016. (http://jmlr.org/papers/v17/15-335.html)
* 31 pages, 5 Figures, technical report
Click to Read Paper and Get Code
With the feature of multi-master bus access, nondestructive contention-based arbitration and flexible configuration, Controller Area Network (CAN) bus is applied into the control system of Wire Harness Assembly Machine (WHAM). To accomplish desired goal, the specific features of the CAN bus is analyzed by compared with other field buses and the functional performances in the CAN bus system of WHAM is discussed. Then the application layer planning of CAN bus for dynamic priority is presented. The critical issue for the use of CAN bus system in WHAM is the data transfer rate between different nodes. So processing efficient model is introduced to assist analyzing data transfer procedure. Through the model, it is convenient to verify the real time feature of the CAN bus system in WHAM.

* 6 pages, 2 figures, 4 tables
Click to Read Paper and Get Code
Online news media sometimes use misleading headlines to lure users to open the news article. These catchy headlines that attract users but disappointed them at the end, are called Clickbaits. Because of the importance of automatic clickbait detection in online medias, lots of machine learning methods were proposed and employed to find the clickbait headlines. In this research, a model using deep learning methods is proposed to find the clickbaits in Clickbait Challenge 2017's dataset. The proposed model gained the first rank in the Clickbait Challenge 2017 in terms of Mean Squared Error. Also, data analytics and visualization techniques are employed to explore and discover the provided dataset to get more insight from the data.

Click to Read Paper and Get Code
Named entity recognition (NER) systems that perform well require task-related and manually annotated datasets. However, they are expensive to develop, and are thus limited in size. As there already exists a large number of NER datasets that share a certain degree of relationship but differ in content, it is important to explore the question of whether such datasets can be combined as a simple method for improving NER performance. To investigate this, we developed a novel locally detecting multitask model using FFNNs. The model relies on encoding variable-length sequences of words into theoretically lossless and unique fixed-size representations. We applied this method to several well-known NER tasks and compared the results of our model to baseline models as well as other published results. As a result, we observed competitive performance in nearly all of the tasks.

* 8 pages, 1 figure, 5 tables (Rejected by ACL2018 with score 3-4-4)
Click to Read Paper and Get Code
In this paper, we explore a new approach to named entity recognition (NER) with the goal of learning from context and fragment features more effectively, contributing to the improvement of overall recognition performance. We use the recent fixed-size ordinally forgetting encoding (FOFE) method to fully encode each sentence fragment and its left-right contexts into a fixed-size representation. Next, we organize the context and fragment features into groups, and feed each feature group to dedicated fully-connected layers. Finally, we merge each group's final dedicated layers and add a shared layer leading to a single output. The outcome of our experiments show that, given only tokenized text and trained word embeddings, our system outperforms our baseline models, and is competitive to the state-of-the-arts of various well-known NER tasks.

* 7 pages, 1 figure, 7 tables (Rejected by EMNLP 2018 with score 3-4-4). arXiv admin note: text overlap with arXiv:1904.03300
Click to Read Paper and Get Code
In this paper, we present our method of using fixed-size ordinally forgetting encoding (FOFE) to solve the word sense disambiguation (WSD) problem. FOFE enables us to encode variable-length sequence of words into a theoretically unique fixed-size representation that can be fed into a feed forward neural network (FFNN), while keeping the positional information between words. In our method, a FOFE-based FFNN is used to train a pseudo language model over unlabelled corpus, then the pre-trained language model is capable of abstracting the surrounding context of polyseme instances in labelled corpus into context embeddings. Next, we take advantage of these context embeddings towards WSD classification. We conducted experiments on several WSD data sets, which demonstrates that our proposed method can achieve comparable performance to that of the state-of-the-art approach at the expense of much lower computational cost.

Click to Read Paper and Get Code
In this paper, we propose several novel deep learning methods for object saliency detection based on the powerful convolutional neural networks. In our approach, we use a gradient descent method to iteratively modify an input image based on the pixel-wise gradients to reduce a cost function measuring the class-specific objectness of the image. The pixel-wise gradients can be efficiently computed using the back-propagation algorithm. The discrepancy between the modified image and the original one may be used as a saliency map for the image. Moreover, we have further proposed several new training methods to learn saliency-specific convolutional nets for object saliency detection, in order to leverage the available pixel-wise segmentation information. Our methods are extremely computationally efficient (processing 20-40 images per second in one GPU). In this work, we use the computed saliency maps for image segmentation. Experimental results on two benchmark tasks, namely Microsoft COCO and Pascal VOC 2012, have shown that our proposed methods can generate high-quality salience maps, clearly outperforming many existing methods. In particular, our approaches excel in handling many difficult images, which contain complex background, highly-variable salient objects, multiple objects, and/or very small salient objects.

* 9 pages, 126 figures, technical report
Click to Read Paper and Get Code