Prior work demonstrated the potential of using the Linear Predictive Coding (LPC) filter to approximate muscle stiffness and damping from computer mouse movements to predict acute stress levels of users. Theoretically, muscle stiffness and damping in the arm can be estimated using a mass-spring-damper (MSD) biomechanical model. However, the damping frequency (i.e., stiffness) and damping ratio values derived using LPC were not yet compared with those from a theoretical MSD model. This work demonstrates that the damping frequency and damping ratio from LPC are significantly correlated with those from an MSD model, thus confirming the validity of using LPC to infer muscle stiffness and damping. We also compare the stress level binary classification performance using the values from LPC and MSD with each other and with neural network-based baselines. We found comparable performance across all conditions demonstrating LPC and MSD model-based stress prediction efficacy, especially for longer mouse trajectories. Clinical relevance: This work demonstrates the validity of the LPC filter to approximate muscle stiffness and damping and predict acute stress from computer mouse movements.
This report describes the submission from Technical University of Catalonia (UPC) to the VoxCeleb Speaker Recognition Challenge (VoxSRC-20) at Interspeech 2020. The final submission is a combination of three systems. System-1 is an autoencoder based approach which tries to reconstruct similar i-vectors, whereas System-2 and -3 are Convolutional Neural Network (CNN) based siamese architectures. The siamese networks have two and three branches, respectively, where each branch is a CNN encoder. The double-branch siamese performs binary classification using cross entropy loss during training. Whereas, our triple-branch siamese is trained to learn speaker embeddings using triplet loss. We provide results of our systems on VoxCeleb-1 test, VoxSRC-20 validation and test sets.
Recently, visual perception (VP) and visual imagery (VI) paradigms are investigated in several brain-computer interface (BCI) studies. VP and VI are defined as a changing of brain signals when perceiving and memorizing visual information, respectively. These paradigms could be alternatives to the previous visual-based paradigms which have limitations such as fatigue and low information transfer rates (ITR). In this study, we analyzed VP and VI to investigate the possibility to control BCI. First, we conducted a time-frequency analysis with event-related spectral perturbation. In addition, two types of decoding accuracies were obtained with convolutional neural network to verify whether the brain signals can be distinguished from each class in the VP and whether they can be differentiated with VP and VI paradigms. As a result, the 6-class classification performance in VP was 32.56% and the binary classification performance which classifies two paradigms was 90.16%.
Aflatoxins are fungal metabolites extensively produced by many different fungal species that may contaminate a wide range of agricultural food products. They have been studied extensively because of being associated with various chronic and acute diseases especially immunosuppression and cancer and their presence in food is strictly monitored and regulated worldwide. Aflatoxin detection and measurement relies mainly on chemical methods usually based on chromatography approaches, and recently developed immunochemical based assays that have advantages but also limitations, since these are expensive and destructive techniques. Nondestructive, optical approaches are recently being developed to assess presence of contamination in a cost and time effective way, maintaining acceptable accuracy and reproducibility. In this paper are presented the results obtained with a simple portable device for nondestructive detection of aflatoxins in almonds. The presented approach is based on the analysis of fluorescence spectra of slurried almonds under 375 nm wavelength excitation. Experiments were conducted with almonds contaminated in the range of 2.7-320.2 ng/g total aflatoxins B (AFB1 + AFB2) as determined by HPLC/FLD. After applying pre-processing steps, spectral analysis was carried out by a binary classification model based on SVM algorithm. A majority vote procedure was then performed on the classification results. In this way we could achieve, as best result, a classification accuracy of 94% (and false negative rate 5%) with a threshold set at 6.4 ng/g. These results illustrate the feasibility of such an approach in the great challenge of aflatoxin detection for food and feed safety.
Current deep learning based detection models tackle detection and segmentation tasks by casting them to pixel or patch-wise classification. To automate the initial mass lesion detection and segmentation on the whole mammographic images and avoid the computational redundancy of patch-based and sliding window approaches, the conditional generative adversarial network (cGAN) was used in this study. Subsequently, feeding the detected regions to the trained densely connected network (DenseNet), the binary classification of benign versus malignant was predicted. We used a combination of publicly available mammographic data repositories to train the pipeline, while evaluating the model's robustness toward our clinically collected repository, which was unseen to the pipeline.
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.
Fragility curves which express the failure probability of a structure, or critical components, as function of a loading intensity measure are nowadays widely used (i) in Seismic Probabilistic Risk Assessment studies, (ii) to evaluate impact of construction details on the structural performance of installations under seismic excitations or under other loading sources such as wind. To avoid the use of parametric models such as lognormal model to estimate fragility curves from a reduced number of numerical calculations, a methodology based on Support Vector Machines coupled with an active learning algorithm is proposed in this paper. In practice, input excitation is reduced to some relevant parameters and, given these parameters, SVMs are used for a binary classification of the structural responses relative to a limit threshold of exceedance. Since the output is not only binary, this is a score, a probabilistic interpretation of the output is exploited to estimate very efficiently fragility curves as score functions or as functions of classical seismic intensity measures.
Objective: Ultrahigh-resolution optical coherence microscopy (OCM) has recently demonstrated its potential for accurate diagnosis of human cervical diseases. One major challenge for clinical adoption, however, is the steep learning curve clinicians need to overcome to interpret OCM images. Developing an intelligent technique for computer-aided diagnosis (CADx) to accurately interpret OCM images will facilitate clinical adoption of the technology and improve patient care. Methods: 497 high-resolution 3-D OCM volumes (600 cross-sectional images each) were collected from 159 ex vivo specimens of 92 female patients. OCM image features were extracted using a convolutional neural network (CNN) model, concatenated with patient information (e.g., age, HPV results), and classified using a support vector machine classifier. Ten-fold cross-validations were utilized to test the performance of the CADx method in a five-class classification task and a binary classification task. Results: An 88.3 plus or minus 4.9% classification accuracy was achieved for five fine-grained classes of cervical tissue, namely normal, ectropion, low-grade and high-grade squamous intraepithelial lesions (LSIL and HSIL), and cancer. In the binary classification task (low-risk [normal, ectropion and LSIL] vs. high-risk [HSIL and cancer]), the CADx method achieved an area-under-the-curve (AUC) value of 0.959 with an 86.7 plus or minus 11.4% sensitivity and 93.5 plus or minus 3.8% specificity. Conclusion: The proposed deep-learning based CADx method outperformed three human experts. It was also able to identify morphological characteristics in OCM images that were consistent with histopathological interpretations. Significance: Label-free OCM imaging, combined with deep-learning based CADx methods, hold a great promise to be used in clinical settings for the effective screening and diagnosis of cervical diseases.
We apply techniques in natural language processing, computational linguistics, and machine-learning to investigate papers in hep-th and four related sections of the arXiv: hep-ph, hep-lat, gr-qc, and math-ph. All of the titles of papers in each of these sections, from the inception of the arXiv until the end of 2017, are extracted and treated as a corpus which we use to train the neural network Word2Vec. A comparative study of common n-grams, linear syntactical identities, word cloud and word similarities is carried out. We find notable scientific and sociological differences between the fields. In conjunction with support vector machines, we also show that the syntactic structure of the titles in different sub-fields of high energy and mathematical physics are sufficiently different that a neural network can perform a binary classification of formal versus phenomenological sections with 87.1% accuracy, and can perform a finer five-fold classification across all sections with 65.1% accuracy.
There is extensive evidence that biological neural networks encode information in the precise timing of the spikes generated and transmitted by neurons, which offers several advantages over rate-based codes. Here we adopt a vector space formulation of spike train sequences and introduce a new liquid state machine (LSM) network architecture and a new forward orthogonal regression algorithm to learn an input-output signal mapping or to decode the brain activity. The proposed algorithm uses precise spike timing to select the presynaptic neurons relevant to each learning task. We show that using precise spike timing to train the LSM and selecting the readout neurons leads to a significant increase in performance on binary classification tasks as well as in decoding neural activity from multielectrode array recordings, compared with what is achieved using the standard architecture and training method.
The WSDM Cup 2017 is a binary classification task for classifying Wikidata revisions into vandalism and non-vandalism. This paper describes our method using some machine learning techniques such as under-sampling, feature selection, stacking and ensembles of models. We confirm the validity of each technique by calculating AUC-ROC of models using such techniques and not using them. Additionally, we analyze the results and gain useful insights into improving models for the vandalism detection task. The AUC-ROC of our final submission after the deadline resulted in 0.94412.
Objective. The main goal of this work is to develop a model for multi-sensor signals such as MEG or EEG signals, that accounts for the inter-trial variability, suitable for corresponding binary classification problems. An important constraint is that the model be simple enough to handle small size and unbalanced datasets, as often encountered in BCI type experiments. Approach. The method involves linear mixed effects statistical model, wavelet transform and spatial filtering, and aims at the characterization of localized discriminant features in multi-sensor signals. After discrete wavelet transform and spatial filtering, a projection onto the relevant wavelet and spatial channels subspaces is used for dimension reduction. The projected signals are then decomposed as the sum of a signal of interest (i.e. discriminant) and background noise, using a very simple Gaussian linear mixed model. Main results. Thanks to the simplicity of the model, the corresponding parameter estimation problem is simplified. Robust estimates of class-covariance matrices are obtained from small sample sizes and an effective Bayes plug-in classifier is derived. The approach is applied to the detection of error potentials in multichannel EEG data, in a very unbalanced situation (detection of rare events). Classification results prove the relevance of the proposed approach in such a context. Significance. The combination of linear mixed model, wavelet transform and spatial filtering for EEG classification is, to the best of our knowledge, an original approach, which is proven to be effective. This paper improves on earlier results on similar problems, and the three main ingredients all play an important role.
In this paper, we explore neural network-based strategies for performing symbol detection in a MIMO-OFDM system. Building on a reservoir computing (RC)-based approach towards symbol detection, we introduce a symmetric and decomposed binary decision neural network to take advantage of the structure knowledge inherent in the MIMO-OFDM system. To be specific, the binary decision neural network is added in the frequency domain utilizing the knowledge of the constellation. We show that the introduced symmetric neural network can decompose the original $M$-ary detection problem into a series of binary classification tasks, thus significantly reducing the neural network detector complexity while offering good generalization performance with limited training overhead. Numerical evaluations demonstrate that the introduced hybrid RC-binary decision detection framework performs close to maximum likelihood model-based symbol detection methods in terms of symbol error rate in the low SNR regime with imperfect channel state information (CSI).
In many real-world pattern recognition scenarios, such as in medical applications, the corresponding classification tasks can be of an imbalanced nature. In the current study, we focus on binary, imbalanced classification tasks, i.e.~binary classification tasks in which one of the two classes is under-represented (minority class) in comparison to the other class (majority class). In the literature, many different approaches have been proposed, such as under- or oversampling, to counter class imbalance. In the current work, we introduce a novel method, which addresses the issues of class imbalance. To this end, we first transfer the binary classification task to an equivalent regression task. Subsequently, we generate a set of negative and positive target labels, such that the corresponding regression task becomes balanced, with respect to the redefined target label set. We evaluate our approach on a number of publicly available data sets in combination with Support Vector Machines. Moreover, we compare our proposed method to one of the most popular oversampling techniques (SMOTE). Based on the detailed discussion of the presented outcomes of our experimental evaluation, we provide promising ideas for future research directions.
One key step in performing quantum machine learning (QML) on noisy intermediate-scale quantum (NISQ) devices is the dimension reduction of the input data prior to their encoding. Traditional principle component analysis (PCA) and neural networks have been used to perform this task; however, the classical and quantum layers are usually trained separately. A framework that allows for a better integration of the two key components is thus highly desirable. Here we introduce a hybrid model combining the quantum-inspired tensor networks (TN) and the variational quantum circuits (VQC) to perform supervised learning tasks, which allows for an end-to-end training. We show that a matrix product state based TN with low bond dimensions performs better than PCA as a feature extractor to compress data for the input of VQCs in the binary classification of MNIST dataset. The architecture is highly adaptable and can easily incorporate extra quantum resource when available.
Most theoretical studies explaining the regularization effect in deep learning have only focused on gradient descent with a sufficient small learning rate or even gradient flow (infinitesimal learning rate). Such researches, however, have neglected a reasonably large learning rate applied in most practical applications. In this work, we characterize the implicit bias effect of deep linear networks for binary classification using the logistic loss in the large learning rate regime, inspired by the seminal work by Lewkowycz et al. [26] in a regression setting with squared loss. They found a learning rate regime with a large stepsize named the catapult phase, where the loss grows at the early stage of training and eventually converges to a minimum that is flatter than those found in the small learning rate regime. We claim that depending on the separation conditions of data, the gradient descent iterates will converge to a flatter minimum in the catapult phase. We rigorously prove this claim under the assumption of degenerate data by overcoming the difficulty of the non-constant Hessian of logistic loss and further characterize the behavior of loss and Hessian for non-separable data. Finally, we demonstrate that flatter minima in the space spanned by non-separable data along with the learning rate in the catapult phase can lead to better generalization empirically.
The backpropagation algorithm is often debated for its biological plausibility. However, various learning methods for neural architecture have been proposed in search of more biologically plausible learning. Most of them have tried to solve the "weight transport problem" and try to propagate errors backward in the architecture via some alternative methods. In this work, we investigated a slightly different approach that uses only the local information which captures spike timing information with no propagation of errors. The proposed learning rule is derived from the concepts of spike timing dependant plasticity and neuronal association. A preliminary evaluation done on the binary classification of MNIST and IRIS datasets with two hidden layers shows comparable performance with backpropagation. The model learned using this method also shows a possibility of better adversarial robustness against the FGSM attack compared to the model learned through backpropagation of cross-entropy loss. The local nature of learning gives a possibility of large scale distributed and parallel learning in the network. And finally, the proposed method is a more biologically sound method that can probably help in understanding how biological neurons learn different abstractions.
Kernel methods play an important role in machine learning applications due to their conceptual simplicity and superior performance on numerous machine learning tasks. Expressivity of a machine learning model, referring to the ability of the model to approximate complex functions, has a significant influence on its performance in these tasks. One approach to enhancing the expressivity of kernel machines is to combine multiple individual kernels to arrive at a more expressive combined kernel. This approach is referred to as multiple kernel learning (MKL). In this work, we propose an MKL method we refer to as quantum MKL, which combines multiple quantum kernels. Our method leverages the power of deterministic quantum computing with one qubit (DQC1) to estimate the combined kernel for a set of classically intractable individual quantum kernels. The combined kernel estimation is achieved without explicitly computing each individual kernel, while still allowing for the tuning of individual kernels in order to achieve better expressivity. Our simulations on two binary classification problems---one performed on a synthetic dataset and the other on a German credit dataset---demonstrate the superiority of the quantum MKL method over single quantum kernel machines.
We prove bounds for the approximation and estimation of certain binary classification functions using ReLU neural networks. Our estimation bounds provide a priori performance guarantees for empirical risk minimization using networks of a suitable size, depending on the number of training samples available. The obtained approximation and estimation rates are independent of the dimension of the input, showing that the curse of dimensionality can be overcome in this setting; in fact, the input dimension only enters in the form of a polynomial factor. Regarding the regularity of the target classification function, we assume the interfaces between the different classes to be locally of Barron-type. We complement our results by studying the relations between various Barron-type spaces that have been proposed in the literature. These spaces differ substantially more from each other than the current literature suggests.
Deep neural networks generalize well despite being exceedingly overparameterized and being trained without explicit regularization. This curious phenomenon has inspired extensive research activity in establishing its statistical principles: Under what conditions is it observed? How do these depend on the data and on the training algorithm? When does regularization benefit generalization? While such questions remain wide open for deep neural nets, recent works have attempted gaining insights by studying simpler, often linear, models. Our paper contributes to this growing line of work by examining binary linear classification under a generative Gaussian mixture model. Motivated by recent results on the implicit bias of gradient descent, we study both max-margin SVM classifiers (corresponding to logistic loss) and min-norm interpolating classifiers (corresponding to least-squares loss). First, we leverage an idea introduced in [V. Muthukumar et al., arXiv:2005.08054, (2020)] to relate the SVM solution to the min-norm interpolating solution. Second, we derive novel non-asymptotic bounds on the classification error of the latter. Combining the two, we present novel sufficient conditions on the covariance spectrum and on the signal-to-noise ratio (SNR) under which interpolating estimators achieve asymptotically optimal performance as overparameterization increases. Interestingly, our results extend to a noisy model with constant probability noise flips. Contrary to previously studied discriminative data models, our results emphasize the crucial role of the SNR and its interplay with the data covariance. Finally, via a combination of analytical arguments and numerical demonstrations we identify conditions under which the interpolating estimator performs better than corresponding regularized estimates.