This paper describes our submission for the SemEval-2019 Suggestion Mining task. A simple Convolutional Neural Network (CNN) classifier with contextual word representations from a pre-trained language model was used for sentence classification. The model is trained using tri-training, a semi-supervised bootstrapping mechanism for labelling unseen data. Tri-training proved to be an effective technique to accommodate domain shift for cross-domain suggestion mining (Subtask B) where there is no hand labelled training data. For in-domain evaluation (Subtask A), we use the same technique to augment the training set. Our system ranks thirteenth in Subtask A with an $F_1$-score of 68.07 and third in Subtask B with an $F_1$-score of 81.94.
Sexism is very common in social media and makes the boundaries of freedom tighter for feminist and female users. There is still no comprehensive classification of sexism attracting natural language processing techniques. Categorizing sexism in social media in the categories of hostile or benevolent sexism are so general that simply ignores the other types of sexism happening in these media. This paper proposes a more comprehensive and in-depth categories of online harassment in social media e.g. twitter into the following categories, "Indirect harassment", "Information threat", "sexual harassment", "Physical harassment" and "Not sexist" and address the challenge of labeling them along with presenting the classification result of the categories. It is preliminary work applying machine learning to learn the concept of sexism and distinguishes itself by looking at more precise categories of sexism in social media.
A growing number of state-of-the-art transfer learning methods employ language models pretrained on large generic corpora. In this paper we present a conceptually simple and effective transfer learning approach that addresses the problem of catastrophic forgetting. Specifically, we combine the task-specific optimization function with an auxiliary language model objective, which is adjusted during the training process. This preserves language regularities captured by language models, while enabling sufficient adaptation for solving the target task. Our method does not require pretraining or finetuning separate components of the network and we train our models end-to-end in a single step. We present results on a variety of challenging affective and text classification tasks, surpassing well established transfer learning methods with greater level of complexity.
Text classification tends to struggle when data is deficient or when it needs to adapt to unseen classes. In such challenging scenarios, recent studies have used meta-learning to simulate the few-shot task, in which new queries are compared to a small support set at the sample-wise level. However, this sample-wise comparison may be severely disturbed by the various expressions in the same class. Therefore, we should be able to learn a general representation of each class in the support set and then compare it to new queries. In this paper, we propose a novel Induction Network to learn such a generalized class-wise representation, by innovatively leveraging the dynamic routing algorithm in meta-learning. In this way, we find the model is able to induce and generalize better. We evaluate the proposed model on a well-studied sentiment classification dataset (English) and a real-world dialogue intent classification dataset (Chinese). Experiment results show that on both datasets, the proposed model significantly outperforms the existing state-of-the-art approaches, proving the effectiveness of class-wise generalization in few-shot text classification.
One of the founding paradigms of machine learning is that a small number of variables is often sufficient to describe high-dimensional data. The minimum number of variables required is called the intrinsic dimension (ID) of the data. Contrary to common intuition, there are cases where the ID varies within the same data set. This fact has been highlighted in technical discussions, but seldom exploited to analyze large data sets and obtain insight into their structure. Here we develop a robust approach to discriminate regions with different local IDs and segment the points accordingly. Our approach is computationally efficient and can be proficiently used even on large data sets. We find that many real-world data sets contain regions with widely heterogeneous dimensions. These regions host points differing in core properties: folded vs unfolded configurations in a protein molecular dynamics trajectory, active vs non-active regions in brain imaging data, and firms with different financial risk in company balance sheets. A simple topological feature, the local ID, is thus sufficient to achieve an unsupervised segmentation of high-dimensional data, complementary to the one given by clustering algorithms.
Neural networks are widely used as a model for classification in a large variety of tasks. Typically, a learnable transformation (i.e. the classifier) is placed at the end of such models returning a value for each class used for classification. This transformation plays an important role in determining how the generated features change during the learning process. In this work we argue that this transformation not only can be fixed (i.e. set as non trainable) with no loss of accuracy, but it can also be used to learn stationary and maximally discriminative embeddings. We show that the stationarity of the embedding and its maximal discriminative representation can be theoretically justified by setting the weights of the fixed classifier to values taken from the coordinate vertices of three regular polytopes available in $\mathbb{R}^d$, namely: the $d$-Simplex, the $d$-Cube and the $d$-Orthoplex. These regular polytopes have the maximal amount of symmetry that can be exploited to generate stationary features angularly centered around their corresponding fixed weights. Our approach improves and broadens the concept of a fixed classifier, recently proposed in \cite{hoffer2018fix}, to a larger class of fixed classifier models. Experimental results confirm both the theoretical analysis and the generalization capability of the proposed method.
Despite their successes in the field of self-learning AI, Convolutional Neural Networks (CNNs) suffer from having too many trainable parameters, impacting computational performance. Several approaches have been proposed to reduce the number of parameters in the visual domain, the Inception architecture [Szegedy et al., 2016] being a prominent example. This raises the question whether the number of trainable parameters in CNNs can also be reduced for 1D inputs, such as time-series data, without incurring a substantial loss in classification performance. We propose and examine two methods for complexity reduction in AstroNet [Shallue & Vanderburg, 2018], a CNN for automatic classification of time-varying brightness data of stars to detect exoplanets. The first method makes only a tactical reduction of layers in AstroNet while the second method also modifies the original input data by means of a Gaussian pyramid. We conducted our experiments with various degrees of dropout regularization. Our results show only a non-substantial loss in accuracy compared to the original AstroNet, while reducing training time up to 85 percent. These results show potential for similar reductions in other CNN applications while largely retaining accuracy.
Channel pruning, which seeks to reduce the model size by removing redundant channels, is a popular solution for deep networks compression. Existing channel pruning methods usually conduct layer-wise channel selection by directly minimizing the reconstruction error of feature maps between the baseline model and the pruned one. However, they ignore the feature and semantic distributions within feature maps and real contribution of channels to the overall performance. In this paper, we propose a new channel pruning method by explicitly using both intermediate outputs of the baseline model and the classification loss of the pruned model to supervise layer-wise channel selection. Particularly, we introduce an additional loss to encode the differences in the feature and semantic distributions within feature maps between the baseline model and the pruned one. By considering the reconstruction error, the additional loss and the classification loss at the same time, our approach can significantly improve the performance of the pruned model. Comprehensive experiments on benchmark datasets demonstrate the effectiveness of the proposed method.
With the rapid development in deep learning, deep neural networks have been widely adopted in many real-life natural language applications. Under deep neural networks, a pre-defined vocabulary is required to vectorize text inputs. The canonical approach to select pre-defined vocabulary is based on the word frequency, where a threshold is selected to cut off the long tail distribution. However, we observed that such simple approach could easily lead to under-sized vocabulary or over-sized vocabulary issues. Therefore, we are interested in understanding how the end-task classification accuracy is related to the vocabulary size and what is the minimum required vocabulary size to achieve a specific performance. In this paper, we provide a more sophisticated variational vocabulary dropout (VVD) based on variational dropout to perform vocabulary selection, which can intelligently select the subset of the vocabulary to achieve the required performance. To evaluate different algorithms on the newly proposed vocabulary selection problem, we propose two new metrics: Area Under Accuracy-Vocab Curve and Vocab Size under X\% Accuracy Drop. Through extensive experiments on various NLP classification tasks, our variational framework is shown to significantly outperform the frequency-based and other selection baselines on these metrics.
Packet classification is a fundamental problem in computer networking. This problem exposes a hard tradeoff between the computation and state complexity, which makes it particularly challenging. To navigate this tradeoff, existing solutions rely on complex hand-tuned heuristics, which are brittle and hard to optimize. In this paper, we propose a deep reinforcement learning (RL) approach to solve the packet classification problem. There are several characteristics that make this problem a good fit for Deep RL. First, many of the existing solutions are iteratively building a decision tree by splitting nodes in the tree. Second, the effects of these actions (e.g., splitting nodes) can only be evaluated once we are done with building the tree. These two characteristics are naturally captured by the ability of RL to take actions that have sparse and delayed rewards. Third, it is computationally efficient to generate data traces and evaluate decision trees, which alleviate the notoriously high sample complexity problem of Deep RL algorithms. Our solution, NeuroCuts, uses succinct representations to encode state and action space, and efficiently explore candidate decision trees to optimize for a global objective. It produces compact decision trees optimized for a specific set of rules and a given performance metric, such as classification time, memory footprint, or a combination of the two. Evaluation on ClassBench shows that NeuroCuts outperforms existing hand-crafted algorithms in classification time by 18% at the median, and reduces both time and memory footprint by up to 3x.
In this paper, we propose a light reflection based face anti-spoofing method named Aurora Guard (AG), which is fast, simple yet effective that has already been deployed in real-world systems serving for millions of users. Specifically, our method first extracts the normal cues via light reflection analysis, and then uses an end-to-end trainable multi-task Convolutional Neural Network (CNN) to not only recover subjects' depth maps to assist liveness classification, but also provide the light CAPTCHA checking mechanism in the regression branch to further improve the system reliability. Moreover, we further collect a large-scale dataset containing $12,000$ live and spoofing samples, which covers abundant imaging qualities and Presentation Attack Instruments (PAI). Extensive experiments on both public and our datasets demonstrate the superiority of our proposed method over the state of the arts.
Epilepsy is a well-known neuronal disorder that can be identified by interpretation of the electroencephalogram (EEG) signal. Usually, the length of an EEG signal is quite long which is challenging to interpret manually. In this work, we propose an automated epileptic seizure detection method by applying a two-step minimization technique: first, we reduce the data points using a statistical sampling technique and then, we minimize the number of features using our novel feature selection algorithm. We then apply different machine learning algorithms for performance measurement of the proposed feature selection algorithm. The experimental results outperform some of the state-of-the-art methods for seizure detection using the reduced data points and the least number of features.
Fine-grained image classification remains challenging due to the large intra-class variance and small inter-class variance. Since the subtle visual differences are only in local regions of discriminative parts among subcategories, part localization is a key issue for fine-grained image classification. Most existing approaches localize object or parts in an image with object or part annotations, which are expensive and labor-consuming. To tackle this issue, we propose a fully unsupervised part mining (UPM) approach to localize the discriminative parts without even image-level annotations, which largely improves the fine-grained classification performance. We first utilize pattern mining techniques to discover frequent patterns, i.e., co-occurrence highlighted regions, in the feature maps extracted from a pre-trained convolutional neural network (CNN) model. Inspired by the fact that these relevant meaningful patterns typically hold appearance and spatial consistency, we then cluster the mined regions to obtain the cluster centers and the discriminative parts surrounding the cluster centers are generated. Importantly, any annotations and sophisticated training procedures are not used in our proposed part localization approach. Finally, a multi-stream classification network is built for aggregating the original, object-level and part-level features simultaneously. Compared with other state-of-the-art approaches, our UPM approach achieves the competitive performance.
In this paper, we develop a new aligned vertex convolutional network model to learn multi-scale local-level vertex features for graph classification. Our idea is to transform the graphs of arbitrary sizes into fixed-sized aligned vertex grid structures, and define a new vertex convolution operation by adopting a set of fixed-sized one-dimensional convolution filters on the grid structure. We show that the proposed model not only integrates the precise structural correspondence information between graphs but also minimises the loss of structural information residing on local-level vertices. Experiments on standard graph datasets demonstrate the effectiveness of the proposed model.
Background: Parkinson's disease (PD) is a prevalent long-term neurodegenerative disease. Though the diagnostic criteria of PD are relatively well defined, the current medical imaging diagnostic procedures are expertise-demanding, and thus call for a higher-integrated AI-based diagnostic algorithm. Methods: In this paper, we proposed an automatic, end-to-end, multi-modality diagnosis framework, including segmentation, registration, feature generation and machine learning, to process the information of the striatum for the diagnosis of PD. Multiple modalities, including T1- weighted MRI and 11C-CFT PET, were used in the proposed framework. The reliability of this framework was then validated on a dataset from the PET center of Huashan Hospital, as the dataset contains paired T1-MRI and CFT-PET images of 18 Normal (NL) subjects and 49 PD subjects. Results: We obtained an accuracy of 100% for the PD/NL classification task, besides, we conducted several comparative experiments to validate the diagnosis ability of our framework. Conclusion: Through experiment we illustrate that (1) automatic segmentation has the same classification effect as the manual segmentation, (2) the multi-modality images generates a better prediction than single modality images, and (3) volume feature is shown to be irrelevant to PD diagnosis.
Most deep learning classification studies assume clean data. However, when dealing with the real world data, we encounter three problems such as 1) missing data, 2) class imbalance, and 3) missing label problems. These problems undermine the performance of a classifier. Various preprocessing techniques have been proposed to mitigate one of these problems, but an algorithm that assumes and resolves all three problems together has not been proposed yet. In this paper, we propose HexaGAN, a generative adversarial network framework that shows promising classification performance for all three problems. We interpret the three problems from a single perspective to solve them jointly. To enable this, the framework consists of six components, which interact with each other. We also devise novel loss functions corresponding to the architecture. The designed loss functions allow us to achieve state-of-the-art imputation performance, with up to a 14% improvement, and to generate high-quality class-conditional data. We evaluate the classification performance (F1-score) of the proposed method with 20% missingness and confirm up to a 5% improvement in comparison with the performance of combinations of state-of-the-art methods.
High performance ultrasonic sensor hardware is mainly used in medical applications. Although, the development in automotive scenarios is towards autonomous driving, the ultrasonic sensor hardware still stays low-cost and low-performance, respectively. To overcome the strict hardware limitations, we propose to use capsule neural networks. By the high classification capability of this network architecture, we can achieve outstanding results for performing a detailed height analysis of detected objects. We apply a novel resorting and reshaping method to feed the neural network with ultrasonic data. This increases classification performance and computation speed. We tested the approach under different environmental conditions to verify that the proposed method is working independent of external parameters that is needed for autonomous driving.
Example synthesis is one of the leading methods to tackle the problem of few-shot learning, where only a small number of samples per class are available. However, current synthesis approaches only address the scenario of a single category label per image. In this work, we propose a novel technique for synthesizing samples with multiple labels for the (yet unhandled) multi-label few-shot classification scenario. We propose to combine pairs of given examples in feature space, so that the resulting synthesized feature vectors will correspond to examples whose label sets are obtained through certain set operations on the label sets of the corresponding input pairs. Thus, our method is capable of producing a sample containing the intersection, union or set-difference of labels present in two input samples. As we show, these set operations generalize to labels unseen during training. This enables performing augmentation on examples of novel categories, thus, facilitating multi-label few-shot classifier learning. We conduct numerous experiments showing promising results for the label-set manipulation capabilities of the proposed approach, both directly (using the classification and retrieval metrics), and in the context of performing data augmentation for multi-label few-shot learning. We propose a benchmark for this new and challenging task and show that our method compares favorably to all the common baselines.
Capturing the meaning of sentences has long been a challenging task. Current models tend to apply linear combinations of word features to conduct semantic composition for bigger-granularity units e.g. phrases, sentences, and documents. However, the semantic linearity does not always hold in human language. For instance, the meaning of the phrase `ivory tower' can not be deduced by linearly combining the meanings of `ivory' and `tower'. To address this issue, we propose a new framework that models different levels of semantic units (e.g. sememe, word, sentence, and semantic abstraction) on a single \textit{Semantic Hilbert Space}, which naturally admits a non-linear semantic composition by means of a complex-valued vector word representation. An end-to-end neural network~\footnote{https://github.com/wabyking/qnn} is proposed to implement the framework in the text classification task, and evaluation results on six benchmarking text classification datasets demonstrate the effectiveness, robustness and self-explanation power of the proposed model. Furthermore, intuitive case studies are conducted to help end users to understand how the framework works.
This paper proposes a data-efficient, semi-supervised, two-pass framework for segmenting bird vocalizations. The framework utilizes a binary classification model to categorize frames of an input audio recording into the background or bird vocalization. The first pass of the framework automatically generates training labels from the input recording itself, while model training and classification is done during the second pass. The proposed framework utilizes a reference directional model for obtaining a feature representation called directional embeddings (DE). This reference directional model acts as an acoustic model for bird vocalizations and is obtained using the mixtures of Von-Mises Fisher distribution (moVMF). The proposed DE space only contains information about bird vocalizations, while no information about the background disturbances is reflected. The framework employs supervised information only for obtaining the reference directional model and avoids the background modeling. Hence, it can be regarded as semi-supervised in nature. The proposed framework is tested on approximately 79000 vocalizations of seven different bird species. The performance of the framework is also analyzed in the presence of noise at different SNRs. Experimental results convey that the proposed framework performs better than the existing bird vocalization segmentation methods.