We propose a learning method well-suited to infer the presence of Tuberculosis (TB) manifestations on Computer Tomography (CT) scans mimicking the radiologist reports. Latent features are extracted from the CT volumes employing the V-Net encoder and those are the input to a Feed-Forward Neural Network (FNN) for multi-class classification. To overtake the issues (e.g., exploding/vanishing gradients, lack of sensibility) that normally appear when training deep 3D models with datasets of limited size and composed of large volumes, our proposal employs: 1) At the network architecture level, the scaled exponential linear unit (SELU) activation which allows the network self-normalization, and 2) at the learning phase, multi-task learning with a loss function weighted by the task homoscedastic uncertainty. The results achieve F1-scores close to or above 0.9 for the detection of TB lesions and a Root Mean Square Error of 1.16 for the number of nodules.
The central idea of this paper is to demonstrate the strength of lyrics for music mining and natural language processing (NLP) tasks using the distributed representation paradigm. For music mining, we address two prediction tasks for songs: genre and popularity. Existing works for both these problems have two major bottlenecks. First, they represent lyrics using handcrafted features that require intricate knowledge of language and music. Second, they consider lyrics as a weak indicator of genre and popularity. We overcome both the bottlenecks by representing lyrics using distributed representation. In our work, genre identification is a multi-class classification task whereas popularity prediction is a binary classification task. We achieve an F1 score of around 0.6 for both the tasks using only lyrics. Distributed representation of words is now heavily used for various NLP algorithms. We show that lyrics can be used to improve the quality of this representation.
This study introduces a method to design a curriculum for machine-learning to maximize the efficiency during the training process of deep neural networks (DNNs) for speech emotion recognition. Previous studies in other machine-learning problems have shown the benefits of training a classifier following a curriculum where samples are gradually presented in increasing level of difficulty. For speech emotion recognition, the challenge is to establish a natural order of difficulty in the training set to create the curriculum. We address this problem by assuming that ambiguous samples for humans are also ambiguous for computers. Speech samples are often annotated by multiple evaluators to account for differences in emotion perception across individuals. While some sentences with clear emotional content are consistently annotated, sentences with more ambiguous emotional content present important disagreement between individual evaluations. We propose to use the disagreement between evaluators as a measure of difficulty for the classification task. We propose metrics that quantify the inter-evaluation agreement to define the curriculum for regression problems and binary and multi-class classification problems. The experimental results consistently show that relying on a curriculum based on agreement between human judgments leads to statistically significant improvements over baselines trained without a curriculum.
Microblog classification has received a lot of attention in recent years. Different classification tasks have been investigated, most of them focusing on classifying microblogs into a small number of classes (five or less) using a training set of manually annotated tweets. Unfortunately, labelling data is tedious and expensive, and finding tweets that cover all the classes of interest is not always straightforward, especially when some of the classes do not frequently arise in practice. In this paper we study an approach to tweet classification based on distant supervision, whereby we automatically transfer labels from one social medium to another for a single-label multi-class classification task. In particular, we apply YouTube video classes to tweets linking to these videos. This provides for free a virtually unlimited number of labelled instances that can be used as training data. The classification experiments we have run show that training a tweet classifier via these automatically labelled data achieves substantially better performance than training the same classifier with a limited amount of manually labelled data; this is advantageous, given that the automatically labelled data come at no cost. Further investigation of our approach shows its robustness when applied with different numbers of classes and across different languages.
Motivation: Novel machine learning and statistical modeling studies rely on standardized comparisons to existing methods using well-studied benchmark datasets. Few tools exist that provide rapid access to many of these datasets through a standardized, user-friendly interface that integrates well with popular data science workflows. Results: This release of PMLB provides the largest collection of diverse, public benchmark datasets for evaluating new machine learning and data science methods aggregated in one location. v1.0 introduces a number of critical improvements developed following discussions with the open-source community. Availability: PMLB is available at https://github.com/EpistasisLab/pmlb. Python and R interfaces for PMLB can be installed through the Python Package Index and Comprehensive R Archive Network, respectively.
Recently, ocular biometrics in unconstrained environments using images obtained at visible wavelength have gained the researchers' attention, especially with images captured by mobile devices. Periocular recognition has been demonstrated to be an alternative when the iris trait is not available due to occlusions or low image resolution. However, the periocular trait does not have the high uniqueness presented in the iris trait. Thus, the use of datasets containing many subjects is essential to assess biometric systems' capacity to extract discriminating information from the periocular region. Also, to address the within-class variability caused by lighting and attributes in the periocular region, it is of paramount importance to use datasets with images of the same subject captured in distinct sessions. As the datasets available in the literature do not present all these factors, in this work, we present a new periocular dataset containing samples from 1,122 subjects, acquired in 3 sessions by 196 different mobile devices. The images were captured under unconstrained environments with just a single instruction to the participants: to place their eyes on a region of interest. We also performed an extensive benchmark with several Convolutional Neural Network (CNN) architectures and models that have been employed in state-of-the-art approaches based on Multi-class Classification, Multitask Learning, Pairwise Filters Network, and Siamese Network. The results achieved in the closed- and open-world protocol, considering the identification and verification tasks, show that this area still needs research and development.
Objective: Target identification in brain-computer interface (BCI) spellers refers to the electroencephalogram (EEG) classification for predicting the target character that the subject intends to spell. When the visual stimulus of each character is tagged with a distinct frequency, the EEG records steady-state visually evoked potentials (SSVEP) whose spectrum is dominated by the harmonics of the target frequency. In this setting, we address the target identification and propose a novel deep neural network (DNN) architecture. Method: The proposed DNN processes the multi-channel SSVEP with convolutions across the sub-bands of harmonics, channels, time, and classifies at the fully connected layer. We test with two publicly available large scale (the benchmark and BETA) datasets consisting of in total 105 subjects with 40 characters. Our first stage training learns a global model by exploiting the statistical commonalities among all subjects, and the second stage fine tunes to each subject separately by exploiting the individualities. Results: Our DNN achieves impressive information transfer rates (ITRs) on both datasets, 265.23 bits/min and 196.59 bits/min, respectively, with only 0.4 seconds of stimulation. The code is available for reproducibility at https://github.com/osmanberke/Deep-SSVEP-BCI. Conclusion: The presented DNN strongly outperforms the state-of-the-art techniques as our accuracy and ITR rates are the highest ever reported performance results on these datasets. Significance: Due to its unprecedentedly high speller ITRs and flawless applicability to general SSVEP systems, our technique has great potential in various biomedical engineering settings of BCIs such as communication, rehabilitation and control.
One of the most rising issues in recent machine learning research is One-Class Classification which considers data set composed of only one class and outliers. It is more reasonable than traditional Multi-Class Classification in dealing with some problematic data set or special cases. Generally, classification accuracy and interpretability for user are considered as trade-off in OCC methods. Classifier based on Hyper-Rectangle (H-RTGL) is a sort of classifier that can be a remedy for such trade-off and uses H-RTGL formulated by conjunction of geometric rules called interval. This interval can be basis of interpretability since it can be easily understood by user. However, existing H-RTGL based OCC classifiers have limitations that (i) most of them cannot reflect density of target class and (ii) that considering density has primitive interval generation method, and (iii) there exists no systematic procedure for hyperparameter of H-RTGL based OCC classifier, which influences classification performance of classifier. Based on these remarks, we suggest One-Class Hyper-Rectangle Descriptor based on density (1-HRD_d) with more elaborate interval generation method including parametric and nonparametric approaches. In addition, we designed Genetic Algorithm (GA) that consists of chromosome structure and genetic operators for systematic generation of 1-HRD_d by optimization of hyperparameter. Our work is validated through a numerical experiment using actual data set with comparison of existing OCC algorithms along with other H-RTGL based classifiers.
A disease that limits a plant from its maximal capacity is defined as plant disease. From the perspective of agriculture, diagnosing plant disease is crucial, as diseases often limit plants' production capacity. However, manual approaches to recognize plant diseases are often temporal, challenging, and time-consuming. Therefore, computerized recognition of plant diseases is highly desired in the field of agricultural automation. Due to the recent improvement of computer vision, identifying diseases using leaf images of a particular plant has already been introduced. Nevertheless, the most introduced model can only diagnose diseases of a specific plant. Hence, in this chapter, we investigate an optimal plant disease identification model combining the diagnosis of multiple plants. Despite relying on multi-class classification, the model inherits a multilabel classification method to identify the plant and the type of disease in parallel. For the experiment and evaluation, we collected data from various online sources that included leaf images of six plants, including tomato, potato, rice, corn, grape, and apple. In our investigation, we implement numerous popular convolutional neural network (CNN) architectures. The experimental results validate that the Xception and DenseNet architectures perform better in multi-label plant disease classification tasks. Through architectural investigation, we imply that skip connections, spatial convolutions, and shorter hidden layer connectivity cause better results in plant disease classification.
We propose a greedy strategy to spectrally train a deep network for multi-class classification. Each layer is defined as a composition of linear weights with the feature map of a Gaussian kernel acting as the activation function. At each layer, the linear weights are learned by maximizing the dependence between the layer output and the labels using the Hilbert Schmidt Independence Criterion (HSIC). By constraining the solution space on the Stiefel Manifold, we demonstrate how our network construct (Kernel Dependence Network or KNet) can be solved spectrally while leveraging the eigenvalues to automatically find the width and the depth of the network. We theoretically guarantee the existence of a solution for the global optimum while providing insight into our network's ability to generalize.
Technology advancements made it easy to measure non-invasive and high-quality electroencephalograph (EEG) signals from human's brain. Hence, development of robust and high-performance AI algorithms becomes crucial to properly process the EEG signals and recognize the patterns, which lead to an appropriate control signal. Despite the advancements in processing the motor imagery EEG signals, the healthcare applications, such as emotion detection, are still in the early stages of AI design. In this paper, we propose a modular framework for the recognition of vowels as the AI part of a brain computer interface system. We carefully designed the modules to discriminate the English vowels given the raw EEG signals, and meanwhile avoid the typical issued with the data-poor environments like most of the healthcare applications. The proposed framework consists of appropriate signal segmentation, filtering, extraction of spectral features, reducing the dimensions by means of principle component analysis, and finally a multi-class classification by decision-tree-based support vector machine (DT-SVM). The performance of our framework was evaluated by a combination of test-set and resubstitution (also known as apparent) error rates. We provide the algorithms of the proposed framework to make it easy for future researchers and developers who want to follow the same workflow.
Recent progress in text classification has been focused on high-resource languages such as English and Chinese. For low-resource languages, amongst them most African languages, the lack of well-annotated data and effective preprocessing, is hindering the progress and the transfer of successful methods. In this paper, we introduce two news datasets (KINNEWS and KIRNEWS) for multi-class classification of news articles in Kinyarwanda and Kirundi, two low-resource African languages. The two languages are mutually intelligible, but while Kinyarwanda has been studied in Natural Language Processing (NLP) to some extent, this work constitutes the first study on Kirundi. Along with the datasets, we provide statistics, guidelines for preprocessing, and monolingual and cross-lingual baseline models. Our experiments show that training embeddings on the relatively higher-resourced Kinyarwanda yields successful cross-lingual transfer to Kirundi. In addition, the design of the created datasets allows for a wider use in NLP beyond text classification in future studies, such as representation learning, cross-lingual learning with more distant languages, or as base for new annotations for tasks such as parsing, POS tagging, and NER. The datasets, stopwords, and pre-trained embeddings are publicly available at https://github.com/Andrews2017/KINNEWS-and-KIRNEWS-Corpus .
Optimal transport (OT) is a powerful geometric tool for comparing two distributions and has been employed in various machine learning applications. In this work, we propose a novel OT formulation that takes feature correlations into account while learning the transport plan between two distributions. We model the feature-feature relationship via a symmetric positive semi-definite Mahalanobis metric in the OT cost function. For a certain class of regularizers on the metric, we show that the optimization strategy can be considerably simplified by exploiting the problem structure. For high-dimensional data, we additionally propose suitable low-dimensional modeling of the Mahalanobis metric. Overall, we view the resulting optimization problem as a non-linear OT problem, which we solve using the Frank-Wolfe algorithm. Empirical results on the discriminative learning setting, such as tag prediction and multi-class classification, illustrate the good performance of our approach.
Cyberbullying is a prevalent social problem that inflicts detrimental consequences to the health and safety of victims such as psychological distress, anti-social behaviour, and suicide. The automation of cyberbullying detection is a recent but widely researched problem, with current research having a strong focus on a binary classification of bullying versus non-bullying. This paper proposes a novel approach to enhancing cyberbullying detection through role modeling. We utilise a dataset from ASKfm to perform multi-class classification to detect participant roles (e.g. victim, harasser). Our preliminary results demonstrate promising performance including 0.83 and 0.76 of F1-score for cyberbullying and role classification respectively, outperforming baselines.
Modern deep learning models have achieved great success in predictive accuracy for many data modalities. However, their application to many real-world tasks is restricted by poor uncertainty estimates, such as overconfidence on out-of-distribution (OOD) data and ungraceful failing under distributional shift. Previous benchmarks have found that ensembles of neural networks (NNs) are typically the best calibrated models on OOD data. Inspired by this, we leverage recent theoretical advances that characterize the function-space prior of an ensemble of infinitely-wide NNs as a Gaussian process, termed the neural network Gaussian process (NNGP). We use the NNGP with a softmax link function to build a probabilistic model for multi-class classification and marginalize over the latent Gaussian outputs to sample from the posterior. This gives us a better understanding of the implicit prior NNs place on function space and allows a direct comparison of the calibration of the NNGP and its finite-width analogue. We also examine the calibration of previous approaches to classification with the NNGP, which treat classification problems as regression to the one-hot labels. In this case the Bayesian posterior is exact, and we compare several heuristics to generate a categorical distribution over classes. We find these methods are well calibrated under distributional shift. Finally, we consider an infinite-width final layer in conjunction with a pre-trained embedding. This replicates the important practical use case of transfer learning and allows scaling to significantly larger datasets. As well as achieving competitive predictive accuracy, this approach is better calibrated than its finite width analogue.
Anomaly detection methods require high-quality features. In recent years, the anomaly detection community has attempted to obtain better features using advances in deep self-supervised feature learning. Surprisingly, a very promising direction, using pretrained deep features, has been mostly overlooked. In this paper, we first empirically establish the perhaps expected, but unreported result, that combining pretrained features with simple anomaly detection and segmentation methods convincingly outperforms, much more complex, state-of-the-art methods. In order to obtain further performance gains in anomaly detection, we adapt pretrained features to the target distribution. Although transfer learning methods are well established in multi-class classification problems, the one-class classification (OCC) setting is not as well explored. It turns out that naive adaptation methods, which typically work well in supervised learning, often result in catastrophic collapse (feature deterioration) and reduce performance in OCC settings. A popular OCC method, DeepSVDD, advocates using specialized architectures, but this limits the adaptation performance gain. We propose two methods for combating collapse: i) a variant of early stopping that dynamically learns the stopping iteration ii) elastic regularization inspired by continual learning. Our method, PANDA, outperforms the state-of-the-art in the OCC, outlier exposure and anomaly segmentation settings by large margins.
Anemia, for which erythropoiesis-stimulating agents (ESAs) and iron supplements (ISs) are used as preventive measures, presents important difficulties for hemodialysis patients. Nevertheless, the number of physicians able to manage such medications appropriately is not keeping pace with the rapid increase of hemodialysis patients. Moreover, the high cost of ESAs imposes heavy burdens on medical insurance systems. An artificial-intelligence-supported anemia control system (AISACS) trained using administration direction data from experienced physicians has been developed by the authors. For the system, appropriate data selection and rectification techniques play important roles. Decision making related to ESAs poses a multi-class classification problem for which a two-step classification technique is introduced. Several validations have demonstrated that AISACS exhibits high performance with correct classification rates of 72-87% and clinically appropriate classification rates of 92-98%.
A Bayesian treatment can mitigate overconfidence in ReLU nets around the training data. But far away from them, ReLU Bayesian neural networks (BNNs) can still underestimate uncertainty and thus be asymptotically overconfident. This issue arises since the output variance of a BNN with finitely many features is quadratic in the distance from the data region. Meanwhile, Bayesian linear models with ReLU features converge, in the infinite-width limit, to a particular Gaussian process (GP) with a variance that grows cubically so that no asymptotic overconfidence can occur. While this may seem of mostly theoretical interest, in this work, we show that it can be used in practice to the benefit of BNNs. We extend finite ReLU BNNs with infinite ReLU features via the GP and show that the resulting model is asymptotically maximally uncertain far away from the data while the BNNs' predictive power is unaffected near the data. Although the resulting model approximates a full GP posterior, thanks to its structure, it can be applied \emph{post-hoc} to any pre-trained ReLU BNN at a low cost.
In forming learning objectives, one oftentimes needs to aggregate a set of individual values to a single output. Such cases occur in the aggregate loss, which combines individual losses of a learning model over each training sample, and in the individual loss for multi-label learning, which combines prediction scores over all class labels. In this work, we introduce the sum of ranked range (SoRR) as a general approach to form learning objectives. A ranked range is a consecutive sequence of sorted values of a set of real numbers. The minimization of SoRR is solved with the difference of convex algorithm (DCA). We explore two applications in machine learning of the minimization of the SoRR framework, namely the AoRR aggregate loss for binary classification and the TKML individual loss for multi-label/multi-class classification. Our empirical results highlight the effectiveness of the proposed optimization framework and demonstrate the applicability of proposed losses using synthetic and real datasets.
Label inventories for fine-grained entity typing have grown in size and complexity. Nonetheless, they exhibit a hierarchical structure. Hyperbolic spaces offer a mathematically appealing approach for learning hierarchical representations of symbolic data. However, it is not clear how to integrate hyperbolic components into downstream tasks. This is the first work that proposes a fully hyperbolic model for multi-class multi-label classification, which performs all operations in hyperbolic space. We evaluate the proposed model on two challenging datasets and compare to different baselines that operate under Euclidean assumptions. Our hyperbolic model infers the latent hierarchy from the class distribution, captures implicit hyponymic relations in the inventory, and shows performance on par with state-of-the-art methods on fine-grained classification with remarkable reduction of the parameter size. A thorough analysis sheds light on the impact of each component in the final prediction and showcases its ease of integration with Euclidean layers.