Wide field small aperture telescopes are widely used for optical transient observations. Detection and classification of astronomical targets in observed images are the most important and basic step. In this paper, we propose an astronomical targets detection and classification framework based on deep neural networks. Our framework adopts the concept of the Faster R-CNN and uses a modified Resnet-50 as backbone network and a Feature Pyramid Network to extract features from images of different astronomical targets. To increase the generalization ability of our framework, we use both simulated and real observation images to train the neural network. After training, the neural network could detect and classify astronomical targets automatically. We test the performance of our framework with simulated data and find that our framework has almost the same detection ability as that of the traditional method for bright and isolated sources and our framework has 2 times better detection ability for dim targets, albeit all celestial objects detected by the traditional method can be classified correctly. We also use our framework to process real observation data and find that our framework can improve 25 % detection ability than that of the traditional method when the threshold of our framework is 0.6. Rapid discovery of transient targets is quite important and we further propose to install our framework in embedded devices such as the Nvidia Jetson Xavier to achieve real-time astronomical targets detection and classification abilities.
Decision trees are popular Classification and Regression tools and, when small-sized, easy to interpret. Traditionally, a greedy approach has been used to build the trees, yielding a very fast training process; however, controlling sparsity (a proxy for interpretability) is challenging. In recent studies, optimal decision trees, where all decisions are optimized simultaneously, have shown a better learning performance, especially when oblique cuts are implemented. In this paper, we propose a continuous optimization approach to build sparse optimal classification trees, based on oblique cuts, with the aim of using fewer predictor variables in the cuts as well as along the whole tree. Both types of sparsity, namely local and global, are modeled by means of regularizations with polyhedral norms. The computational experience reported supports the usefulness of our methodology. In all our data sets, local and global sparsity can be improved without harming classification accuracy. Unlike greedy approaches, our ability to easily trade in some of our classification accuracy for a gain in global sparsity is shown.
In some applications, acquiring covariates comes at a cost which is not negligible. For example in the medical domain, in order to classify whether a patient has diabetes or not, measuring glucose tolerance can be expensive. Assuming that the cost of each covariate, and the cost of misclassification can be specified by the user, our goal is to minimize the (expected) total cost of classification, i.e. the cost of misclassification plus the cost of the acquired covariates. We formalize this optimization goal using the (conditional) Bayes risk and describe the optimal solution using a recursive procedure. Since the procedure is computationally infeasible, we consequently introduce two assumptions: (1) the optimal classifier can be represented by a generalized additive model, (2) the optimal sets of covariates are limited to a sequence of sets of increasing size. We show that under these two assumptions, a computationally efficient solution exists. Furthermore, on several medical datasets, we show that the proposed method achieves in most situations the lowest total costs when compared to various previous methods. Finally, we weaken the requirement on the user to specify all misclassification costs by allowing the user to specify the minimally acceptable recall (target recall). Our experiments confirm that the proposed method achieves the target recall while minimizing the false discovery rate and the covariate acquisition costs better than previous methods.
We consider the problem of learning optimal binary classification trees. Literature on the topic has burgeoned in recent years, motivated both by the empirical suboptimality of heuristic approaches and the tremendous improvements in mixed-integer programming (MIP) technology. Yet, existing approaches from the literature do not leverage the power of MIP to its full extent. Indeed, they rely on weak formulations, resulting in slow convergence and large optimality gaps. To fill this gap in the literature, we propose a flow-based MIP formulation for optimal binary classification trees that has a stronger linear programming relaxation. Our formulation presents an attractive decomposable structure. We exploit this structure and max-flow/min-cut duality to derive a Benders' decomposition method, which scales to larger instances. We conduct extensive computational experiments on standard benchmark datasets on which we show that our proposed approaches are 50 times faster than state-of-the art MIP-based techniques and improve out of sample performance up to 13.8%.
Diagnosis of breast cancer has been well studied in the past. Multiple linear programming models have been devised to approximate the relationship between cell features and tumour malignancy. However, these models are less capable in handling non-linear correlations. Neural networks instead are powerful in processing complex non-linear correlations. It is thus certainly beneficial to approach this cancer diagnosis problem with a model based on neural network. Particularly, introducing bias to neural network training process is deemed as an important means to increase training efficiency. Out of a number of popular proposed methods for introducing artificial bias, Bimodal Distribution Removal (BDR) presents ideal efficiency improvement results and fair simplicity in implementation. However, this paper examines the effectiveness of BDR against the target cancer diagnosis classification problem and shows that BDR process in fact negatively impacts classification performance. In addition, this paper also explores genetic algorithm as an efficient tool for feature selection and produced significantly better results comparing to baseline model that without any feature selection in place
While deep learning strategies achieve outstanding results in computer vision tasks, one issue remains: The current strategies rely heavily on a huge amount of labeled data. In many real-world problems, it is not feasible to create such an amount of labeled training data. Therefore, it is common to incorporate unlabeled data into the training process to reach equal results with fewer labels. Due to a lot of concurrent research, it is difficult to keep track of recent developments. In this survey, we provide an overview of often used ideas and methods in image classification with fewer labels. We compare 34 methods in detail based on their performance and their commonly used ideas rather than a fine-grained taxonomy. In our analysis, we identify three major trends that lead to future research opportunities. 1. State-of-the-art methods are scaleable to real-world applications in theory but issues like class imbalance, robustness, or fuzzy labels are not considered. 2. The degree of supervision which is needed to achieve comparable results to the usage of all labels is decreasing and therefore methods need to be extended to settings with a variable number of classes. 3. All methods share some common ideas but we identify clusters of methods that do not share many ideas. We show that combining ideas from different clusters can lead to better performance.
A coreset is a subset of the training set, using which a machine learning algorithm obtains performances similar to what it would deliver if trained over the whole original data. Coreset discovery is an active and open line of research as it allows improving training speed for the algorithms and may help human understanding the results. Building on previous works, a novel approach is presented: candidate corsets are iteratively optimized, adding and removing samples. As there is an obvious trade-off between limiting training size and quality of the results, a multi-objective evolutionary algorithm is used to minimize simultaneously the number of points in the set and the classification error. Experimental results on non-trivial benchmarks show that the proposed approach is able to deliver results that allow a classifier to obtain lower error and better ability of generalizing on unseen data than state-of-the-art coreset discovery techniques.
Recently, Nystr\"{o}m method has proved its prominence empirically and theoretically in speeding up the training of kernel machines while retaining satisfactory performances and accuracy. So far, there are several different approaches proposed to exploit Nystr\"{o}m method in scaling up kernel machines. However, there is no comparative study over these approaches, and they were individually analyzed for specific types of kernel machines. Therefore, it remains a question that the philosophy of which approach is more promising when it extends to other kernel machines. In this work, motivated by the column inclusion property of Gram matrices, we develop a new approach with a clear geometric interpretation for running Nystr\"{o}m-based kernel machines. We show that the other two well-studied approaches can be equivalently transformed to be our proposed one. Consequently, analysis established for the proposed approach also works for these two. Particularly, our proposed approach makes it possible to develop approximation errors in a general setting. Besides, our analysis also manifests the relations among the aforementioned two approaches and another naive one. First, the analytical forms of the corresponding approximate solutions are only at odds with one term. Second, the naive approach can be implemented efficiently by sharing the same training procedure with others. These analytical results lead to the conjecture that the naive approach can provide more accurate approximate solutions than the other two sophisticated approaches. Since our analysis also offers ways for computing the accuracy of these approximate solutions, we run experiments with classification tasks to confirm our conjecture.
On 21-22 November 2019, about 30 researchers gathered in Victoria, BC, Canada, for the workshop "Detection and Classification in Marine Bioacoustics with Deep Learning" organized by MERIDIAN and hosted by Ocean Networks Canada. The workshop was attended by marine biologists, data scientists, and computer scientists coming from both Canadian coasts and the US and representing a wide spectrum of research organizations including universities, government (Fisheries and Oceans Canada, National Oceanic and Atmospheric Administration), industry (JASCO Applied Sciences, Google, Axiom Data Science), and non-for-profits (Orcasound, OrcaLab). Consisting of a mix of oral presentations, open discussion sessions, and hands-on tutorials, the workshop program offered a rare opportunity for specialists from distinctly different domains to engage in conversation about deep learning and its promising potential for the development of detection and classification algorithms in underwater acoustics. In this workshop report, we summarize key points from the presentations and discussion sessions.
Graph convolutional neural networks, which learn aggregations over neighbor nodes, have achieved great performance in node classification tasks. However, recent studies reported that such graph convolutional node classifier can be deceived by adversarial perturbations on graphs. Abusing graph convolutions, a node's classification result can be influenced by poisoning its neighbors. Given an attributed graph and a node classifier, how can we evaluate robustness against such indirect adversarial attacks? Can we generate strong adversarial perturbations which are effective on not only one-hop neighbors, but more far from the target? In this paper, we demonstrate that the node classifier can be deceived with high-confidence by poisoning just a single node even two-hops or more far from the target. Towards achieving the attack, we propose a new approach which searches smaller perturbations on just a single node far from the target. In our experiments, our proposed method shows 99% attack success rate within two-hops from the target in two datasets. We also demonstrate that m-layer graph convolutional neural networks have chance to be deceived by our indirect attack within m-hop neighbors. The proposed attack can be used as a benchmark in future defense attempts to develop graph convolutional neural networks with having adversary robustness.
Machine learning models provide statistically impressive results which might be individually unreliable. To provide reliability, we propose an Epistemic Classifier (EC) that can provide justification of its belief using support from the training dataset as well as quality of reconstruction. Our approach is based on modified variational auto-encoders that can identify a semantically meaningful low-dimensional space where perceptually similar instances are close in $\ell_2$-distance too. Our results demonstrate improved reliability of predictions and robust identification of samples with adversarial attacks as compared to baseline of softmax-based thresholding.
Documents are composed of smaller pieces - paragraphs, sentences, and tokens - that have complex relationships between one another. Sentiment classification models that take into account the structure inherent in these documents have a theoretical advantage over those that do not. At the same time, transfer learning models based on language model pretraining have shown promise for document classification. However, these two paradigms have not been systematically compared and it is not clear under which circumstances one approach is better than the other. In this work we empirically compare hierarchical models and transfer learning for document-level sentiment classification. We show that non-trivial hierarchical models outperform previous baselines and transfer learning on document-level sentiment classification in five languages.
Pathology slides of lung malignancies are classified using resource-frugal convolution neural networks (CNNs) that may be deployed on mobile devices. In particular, the challenging task of distinguishing adenocarcinoma (LUAD) and squamous-cell carcinoma (LUSC) lung cancer subtypes is approached in two stages. First, whole-slide histopathology images are downsampled to a size too large for CNN analysis but large enough to retain key anatomic detail. The downsampled images are decomposed into smaller square tiles, which are sifted based on their image entropies. A lightweight CNN produces tile-level classifications that are aggregated to classify the slide. The resulting accuracies are comparable to those obtained with much more complex CNNs and larger training sets. To allow clinicians to visually assess the basis for the classification -- that is, to see the image regions that underlie it -- color-coded probability maps are created by overlapping tiles and averaging the tile-level probabilities at a pixel level.
In this project, we aim to classify the speech taken as one of the four emotions namely, sadness, anger, fear and happiness. The samples that have been taken to complete this project are taken from Linguistic Data Consortium (LDC) and UGA database. The important characteristics determined from the samples are energy, pitch, MFCC coefficients, LPCC coefficients and speaker rate. The classifier used to classify these emotional states is Support Vector Machine (SVM) and this is done using two classification strategies: One against All (OAA) and Gender Dependent Classification. Furthermore, a comparative analysis has been conducted between the two and LPCC and MFCC algorithms as well.
Multi-field packet classification is a crucial component in modern software-defined data center networks. To achieve high throughput and low latency, state-of-the-art algorithms strive to fit the rule lookup data structures into on-die caches; however, they do not scale well with the number of rules. We present a novel approach, NuevoMatch, which improves the memory scaling of existing methods. A new data structure, Range Query Recursive Model Index (RQ-RMI), is the key component that enables NuevoMatch to replace most of the accesses to main memory with model inference computations. We describe an efficient training algorithm that guarantees the correctness of the RQ-RMI-based classification. The use of RQ-RMI allows the rules to be compressed into model weights that fit into the hardware cache. Further, it takes advantage of the growing support for fast neural network processing in modern CPUs, such as wide vector instructions, achieving a rate of tens of nanoseconds per lookup. Our evaluation using 500K multi-field rules from the standard ClassBench benchmark shows a geometric mean compression factor of 4.9x, 8x, and 82x, and average performance improvement of 2.4x, 2.6x, and 1.6x in throughput compared to CutSplit, NeuroCuts, and TupleMerge, all state-of-the-art algorithms.
The support vector machine (SVM) and deep learning (e.g., convolutional neural networks (CNNs)) are the two most famous algorithms in small and big data, respectively. Nonetheless, smaller datasets may be very important, costly, and not easy to obtain in a short time. This paper proposes a novel convolutional SVM (CSVM) that has both advantages of CNN and SVM to improve the accuracy and effectiveness of mining smaller datasets. The proposed CSVM adapts the convolution product from CNN to learn new information hidden deeply in the datasets. In addition, it uses a modified simplified swarm optimization (SSO) to help train the CSVM to update classifiers, and then the traditional SVM is implemented as the fitness for the SSO to estimate the accuracy. To evaluate the performance of the proposed CSVM, experiments were conducted to test five well-known benchmark databases for the classification problem. Numerical experiments compared favorably with those obtained using SVM, 3-layer artificial NN (ANN), and 4-layer ANN. The results of these experiments verify that the proposed CSVM with the proposed SSO can effectively increase classification accuracy.
A comprehensive and high-quality lexicon plays a crucial role in traditional text classification approaches. And it improves the utilization of the linguistic knowledge. Although it is helpful for the task, the lexicon has got little attention in recent neural network models. Firstly, getting a high-quality lexicon is not easy. We lack an effective automated lexicon extraction method, and most lexicons are hand crafted, which is very inefficient for big data. What's more, there is no an effective way to use a lexicon in a neural network. To address those limitations, we propose a Pre-Attention mechanism for text classification in this paper, which can learn attention of different words according to their effects in the classification tasks. The words with different attention can form a domain lexicon. Experiments on three benchmark text classification tasks show that our models get competitive result comparing with the state-of-the-art methods. We get 90.5% accuracy on Stanford Large Movie Review dataset, 82.3% on Subjectivity dataset, 93.7% on Movie Reviews. And compared with the text classification model without Pre-Attention mechanism, those with Pre-Attention mechanism improve by 0.9%-2.4% accuracy, which proves the validity of the Pre-Attention mechanism. In addition, the Pre-Attention mechanism performs well followed by different types of neural networks (e.g., convolutional neural networks and Long Short-Term Memory networks). For the same dataset, when we use Pre-Attention mechanism to get attention value followed by different neural networks, those words with high attention values have a high degree of coincidence, which proves the versatility and portability of the Pre-Attention mechanism. we can get stable lexicons by attention values, which is an inspiring method of information extraction.
Graph neural networks (GNNs) have received much attention recently because of their excellent performance on graph-based tasks. However, existing research on GNNs focuses on designing more effective models without considering much about the quality of the input data. In this paper, we propose self-enhanced GNN (SEG), which improves the quality of the input data using the outputs of existing GNN models for better performance on semi-supervised node classification. As graph data consist of both topology and node labels, we improve input data quality from both perspectives. For topology, we observe that higher classification accuracy can be achieved when the ratio of inter-class edges (connecting nodes from different classes) is low and propose topology update to remove inter-class edges and add intra-class edges. For node labels, we propose training node augmentation, which enlarges the training set using the labels predicted by existing GNN models. SEG is a general framework that can be easily combined with existing GNN models. Experimental results validate that SEG consistently improves the performance of well-known GNN models such as GCN, GAT and SGC across different datasets.
In this paper, the task of cross-network node classification, which leverages the abundant labeled nodes from a source network to help classify unlabeled nodes in a target network, is studied. The existing domain adaptation algorithms generally fail to model the network structural information, and the current network embedding models mainly focus on single-network applications. Thus, both of them cannot be directly applied to solve the cross-network node classification problem. This motivates us to propose an adversarial cross-network deep network embedding (ACDNE) model to integrate adversarial domain adaptation with deep network embedding so as to learn network-invariant node representations that can also well preserve the network structural information. In ACDNE, the deep network embedding module utilizes two feature extractors to jointly preserve attributed affinity and topological proximities between nodes. In addition, a node classifier is incorporated to make node representations label-discriminative. Moreover, an adversarial domain adaptation technique is employed to make node representations network-invariant. Extensive experimental results demonstrate that the proposed ACDNE model achieves the state-of-the-art performance in cross-network node classification.
As experimental efforts are costly and time consuming, computational characterization of enzyme capabilities is an attractive alternative. We present and evaluate several machine-learning models to predict which of 983 distinct enzymes, as defined via the Enzyme Commission, EC, numbers, are likely to interact with a given query molecule. Our data consists of enzyme-substrate interactions from the BRENDA database. Some interactions are attributed to natural selection and involve the enzyme's natural substrates. The majority of the interactions however involve non-natural substrates, thus reflecting promiscuous enzymatic activities. We frame this enzyme promiscuity prediction problem as a multi-label classification task. We maximally utilize inhibitor and unlabelled data to train prediction models that can take advantage of known hierarchical relationships between enzyme classes. We report that a hierarchical multi-label neural network, EPP-HMCNF, is the best model for solving this problem, outperforming k-nearest neighbors similarity-based and other machine learning models. We show that inhibitor information during training consistently improves predictive power, particularly for EPP-HMCNF. We also show that all promiscuity prediction models perform worse under a realistic data split when compared to a random data split, and when evaluating performance on non-natural substrates compared to natural substrates. We provide Python code for EPP-HMCNF and other models in a repository termed EPP (Enzyme Promiscuity Prediction) at https://github.com/hassounlab/EPP.