The clinical spectrum of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the strain of coronavirus that caused the COVID-19 pandemic, is broad, extending from asymptomatic infection to severe immunopulmonary reactions that, if not categorized properly, may be life-threatening. Researchers rate COVID-19 patients on a scale from 1 to 8 according to the severity level of COVID-19, 1 being healthy and 8 being extremely sick, based on a multitude of factors including number of clinic visits, days since the first sign of symptoms, and more. However, there are two issues with the current state of severity level designation. Firstly, there exists variation among researchers in determining these patient scores, which may lead to improper treatment. Secondly, researchers use a variety of metrics to determine patient severity level, including metrics involving plasma collection that require invasive procedures. This project aims to remedy both issues by introducing a machine learning framework that unifies severity level designations based on noninvasive saliva biomarkers. Our results show that we can successfully use machine learning on salivaomics data to predict the severity level of COVID-19 patients, indicating the presence of viral load using saliva biomarkers.
Understanding non-linear relationships among financial instruments has various applications in investment processes ranging from risk management, portfolio construction and trading strategies. Here, we focus on interconnectedness among stocks based on their correlation matrix which we represent as a network with the nodes representing individual stocks and the weighted links between pairs of nodes representing the corresponding pair-wise correlation coefficients. The traditional network science techniques, which are extensively utilized in financial literature, require handcrafted features such as centrality measures to understand such correlation networks. However, manually enlisting all such handcrafted features may quickly turn out to be a daunting task. Instead, we propose a new approach for studying nuances and relationships within the correlation network in an algorithmic way using a graph machine learning algorithm called Node2Vec. In particular, the algorithm compresses the network into a lower dimensional continuous space, called an embedding, where pairs of nodes that are identified as similar by the algorithm are placed closer to each other. By using log returns of S&P 500 stock data, we show that our proposed algorithm can learn such an embedding from its correlation network. We define various domain specific quantitative (and objective) and qualitative metrics that are inspired by metrics used in the field of Natural Language Processing (NLP) to evaluate the embeddings in order to identify the optimal one. Further, we discuss various applications of the embeddings in investment management.
We introduce a novel machine learning strategy, kernel addition Gaussian process regression (KA-GPR), in molecular-orbital-based machine learning (MOB-ML) to learn the total correlation energies of general electronic structure theories for closed- and open-shell systems by introducing a machine learning strategy. The learning efficiency of MOB-ML (KA-GPR) is the same as the original MOB-ML method for the smallest criegee molecule, which is a closed-shell molecule with multi-reference characters. In addition, the prediction accuracies of different small free radicals could reach the chemical accuracy of 1 kcal/mol by training on one example structure. Accurate potential energy surfaces for the H10 chain (closed-shell) and water OH bond dissociation (open-shell) could also be generated by MOB-ML (KA-GPR). To explore the breadth of chemical systems that KA-GPR can describe, we further apply MOB-ML to accurately predict the large benchmark datasets for closed- (QM9, QM7b-T, GDB-13-T) and open-shell (QMSpin) molecules.
Applications of machine learning (ML) are growing by the day for many unique and challenging scientific applications. However, a crucial challenge facing these applications is their need for ultra low-latency and on-detector ML capabilities. Given the slowdown in Moore's law and Dennard scaling, coupled with the rapid advances in scientific instrumentation that is resulting in growing data rates, there is a need for ultra-fast ML at the extreme edge. Fast ML at the edge is essential for reducing and filtering scientific data in real-time to accelerate science experimentation and enable more profound insights. To accelerate real-time scientific edge ML hardware and software solutions, we need well-constrained benchmark tasks with enough specifications to be generically applicable and accessible. These benchmarks can guide the design of future edge ML hardware for scientific applications capable of meeting the nanosecond and microsecond level latency requirements. To this end, we present an initial set of scientific ML benchmarks, covering a variety of ML and embedded system techniques.
We suggest a novel classification algorithm that is based on local approximations and explain its connections with Artificial Neural Networks (ANNs) and Nearest Neighbour classifiers. We illustrate it on the datasets MNIST and EMNIST of images of handwritten digits. We use the dataset MNIST to find parameters of our algorithm and apply it with these parameters to the challenging EMNIST dataset. It is demonstrated that the algorithm misclassifies 0.42% of the images of EMNIST and therefore significantly outperforms predictions by humans and shallow artificial neural networks (ANNs with few hidden layers) that both have more than 1.3% of errors
Recent works have revealed an essential paradigm in designing loss functions that differentiate individual losses vs. aggregate losses. The individual loss measures the quality of the model on a sample, while the aggregate loss combines individual losses/scores over each training sample. Both have a common procedure that aggregates a set of individual values to a single numerical value. The ranking order reflects the most fundamental relation among individual values in designing losses. In addition, decomposability, in which a loss can be decomposed into an ensemble of individual terms, becomes a significant property of organizing losses/scores. This survey provides a systematic and comprehensive review of rank-based decomposable losses in machine learning. Specifically, we provide a new taxonomy of loss functions that follows the perspectives of aggregate loss and individual loss. We identify the aggregator to form such losses, which are examples of set functions. We organize the rank-based decomposable losses into eight categories. Following these categories, we review the literature on rank-based aggregate losses and rank-based individual losses. We describe general formulas for these losses and connect them with existing research topics. We also suggest future research directions spanning unexplored, remaining, and emerging issues in rank-based decomposable losses.
The improvement of traffic efficiency at urban intersections receives strong research interest in the field of automated intersection management. So far, mostly non-learning algorithms like reservation or optimization-based ones were proposed to solve the underlying multi-agent planning problem. At the same time, automated driving functions for a single ego vehicle are increasingly implemented using machine learning methods. In this work, we build upon a previously presented graph-based scene representation and graph neural network to approach the problem using reinforcement learning. The scene representation is improved in key aspects by using edge features in addition to the existing node features for the vehicles. This leads to an increased representation quality that is leveraged by an updated network architecture. The paper provides an in-depth evaluation of the proposed method against baselines that are commonly used in automatic intersection management. Compared to a traditional signalized intersection and an enhanced first-in-first-out scheme, a significant reduction of induced delay is observed at varying traffic densities. Finally, the generalization capability of the graph-based representation is evaluated by testing the policy on intersection layouts not seen during training. The model generalizes virtually without restrictions to smaller intersection layouts and within certain limits to larger ones.
Over the last few years, research in automatic sleep scoring has mainly focused on developing increasingly complex deep learning architectures. However, recently these approaches achieved only marginal improvements, often at the expense of requiring more data and more expensive training procedures. Despite all these efforts and their satisfactory performance, automatic sleep staging solutions are not widely adopted in a clinical context yet. We argue that most deep learning solutions for sleep scoring are limited in their real-world applicability as they are hard to train, deploy, and reproduce. Moreover, these solutions lack interpretability and transparency, which are often key to increase adoption rates. In this work, we revisit the problem of sleep stage classification using classical machine learning. Results show that competitive performance can be achieved with a conventional machine learning pipeline consisting of preprocessing, feature extraction, and a simple machine learning model. In particular, we analyze the performance of a linear model and a non-linear (gradient boosting) model. Our approach surpasses state-of-the-art (that uses the same data) on two public datasets: Sleep-EDF SC-20 (MF1 0.810) and Sleep-EDF ST (MF1 0.795), while achieving competitive results on Sleep-EDF SC-78 (MF1 0.775) and MASS SS3 (MF1 0.817). We show that, for the sleep stage scoring task, the expressiveness of an engineered feature vector is on par with the internally learned representations of deep learning models. This observation opens the door to clinical adoption, as a representative feature vector allows to leverage both the interpretability and successful track record of traditional machine learning models.
The recently increased complexity of Machine Learning (ML) methods, led to the necessity to lighten both the research and industry development processes. ML pipelines have become an essential tool for experts of many domains, data scientists and researchers, allowing them to easily put together several ML models to cover the full analytic process starting from raw datasets. Over the years, several solutions have been proposed to automate the building of ML pipelines, most of them focused on semantic aspects and characteristics of the input dataset. However, an approach taking into account the new quality concerns needed by ML systems (like fairness, interpretability, privacy, etc.) is still missing. In this paper, we first identify, from the literature, key quality attributes of ML systems. Further, we propose a new engineering approach for quality ML pipeline by properly extending the Feature Models meta-model. The presented approach allows to model ML pipelines, their quality requirements (on the whole pipeline and on single phases), and quality characteristics of algorithms used to implement each pipeline phase. Finally, we demonstrate the expressiveness of our model considering the classification problem.
We introduce a framework of the equivariant convolutional quantum algorithms which is tailored for a number of machine-learning tasks on physical systems with arbitrary SU$(d)$ symmetries. It allows us to enhance a natural model of quantum computation -- permutational quantum computing (PQC) [Quantum Inf. Comput., 10, 470-497 (2010)] -- and define a more powerful model: PQC+. While PQC was shown to be efficiently classically simulatable, we exhibit a problem which can be efficiently solved on PQC+ machine, whereas no classical polynomial time algorithm is known; thus providing evidence against PQC+ being classically simulatable. We further discuss practical quantum machine learning algorithms which can be carried out in the paradigm of PQC+.
The electrocardiogram (ECG) is routinely used in hospitals to analyze cardiovascular status and health of an individual. Abnormal heart rhythms can be a precursor to more serious conditions including sudden cardiac death. Classifying abnormal rhythms is a laborious process prone to error. Therefore, tools that perform automated classification with high accuracy are highly desirable. The work presented classifies five different types of ECG arrhythmia based on AAMI EC57 standard and using the MIT-BIH data set. These include non-ectopic (normal), supraventricular, ventricular, fusion, and unknown beat. By appropriately transforming pre-processed ECG waveforms into a rich feature space along with appropriate post-processing and utilizing deep convolutional neural networks post fine-tuning and hyperparameter selection, it is shown that highly accurate classification for the five waveform types can be obtained. Performance on the test set indicated higher overall accuracy (98.62%), as well as better performance in classifying each of the five waveforms than hitherto reported in literature.
This paper provides a comprehensive survey of bias mitigation methods for achieving fairness in Machine Learning (ML) models. We collect a total of 341 publications concerning bias mitigation for ML classifiers. These methods can be distinguished based on their intervention procedure (i.e., pre-processing, in-processing, post-processing) and the technique they apply. We investigate how existing bias mitigation methods are evaluated in the literature. In particular, we consider datasets, metrics and benchmarking. Based on the gathered insights (e.g., What is the most popular fairness metric? How many datasets are used for evaluating bias mitigation methods?), we hope to support practitioners in making informed choices when developing and evaluating new bias mitigation methods.
Low-order functional ANOVA (fANOVA) models have been rediscovered in the machine learning (ML) community under the guise of inherently interpretable machine learning. Explainable Boosting Machines or EBM (Lou et al. 2013) and GAMI-Net (Yang et al. 2021) are two recently proposed ML algorithms for fitting functional main effects and second-order interactions. We propose a new algorithm, called GAMI-Tree, that is similar to EBM, but has a number of features that lead to better performance. It uses model-based trees as base learners and incorporates a new interaction filtering method that is better at capturing the underlying interactions. In addition, our iterative training method converges to a model with better predictive performance, and the embedded purification ensures that interactions are hierarchically orthogonal to main effects. The algorithm does not need extensive tuning, and our implementation is fast and efficient. We use simulated and real datasets to compare the performance and interpretability of GAMI-Tree with EBM and GAMI-Net.
Clustering is an unsupervised machine learning methodology where unlabeled elements/objects are grouped together aiming to the construction of well-established clusters that their elements are classified according to their similarity. The goal of this process is to provide a useful aid to the researcher that will help her/him to identify patterns among the data. Dealing with large databases, such patterns may not be easily detectable without the contribution of a clustering algorithm. This article provides a deep description of the most widely used clustering methodologies accompanied by useful presentations concerning suitable parameter selection and initializations. Simultaneously, this article not only represents a review highlighting the major elements of examined clustering techniques but emphasizes the comparison of these algorithms' clustering efficiency based on 3 datasets, revealing their existing weaknesses and capabilities through accuracy and complexity, during the confrontation of discrete and continuous observations. The produced results help us extract valuable conclusions about the appropriateness of the examined clustering techniques in accordance with the dataset's size.
Hyperparameter optimization (HPO) and neural architecture search (NAS) are methods of choice to obtain the best-in-class machine learning models, but in practice they can be costly to run. When models are trained on large datasets, tuning them with HPO or NAS rapidly becomes prohibitively expensive for practitioners, even when efficient multi-fidelity methods are employed. We propose an approach to tackle the challenge of tuning machine learning models trained on large datasets with limited computational resources. Our approach, named PASHA, extends ASHA and is able to dynamically allocate maximum resources for the tuning procedure depending on the need. The experimental comparison shows that PASHA identifies well-performing hyperparameter configurations and architectures while consuming significantly fewer computational resources than ASHA.
Business analytics and machine learning have become essential success factors for various industries - with the downside of cost-intensive gathering and labeling of data. Few-shot learning addresses this challenge and reduces data gathering and labeling costs by learning novel classes with very few labeled data. In this paper, we design a human-in-the-loop (HITL) system for few-shot learning and analyze an extensive range of mechanisms that can be used to acquire human expert knowledge for instances that have an uncertain prediction outcome. We show that the acquisition of human expert knowledge significantly accelerates the few-shot model performance given a negligible labeling effort. We validate our findings in various experiments on a benchmark dataset in computer vision and real-world datasets. We further demonstrate the cost-effectiveness of HITL systems for few-shot learning. Overall, our work aims at supporting researchers and practitioners in effectively adapting machine learning models to novel classes at reduced costs.
The growing adoption of IoT devices for healthcare has enabled researchers to build intelligence using all the data produced by these devices. Monitoring and diagnosing health have been the two most common scenarios where such devices have proven beneficial. Achieving high prediction accuracy was a top priority initially, but the focus has slowly shifted to efficiency and higher throughput, and processing the data from these devices in a distributed manner has proven to help achieve both. Since the field of machine learning is vast with numerous state-of-the-art algorithms in play, it has been a challenge to identify the algorithms that perform best in different scenarios. In this literature review, we explored the distributed machine learning algorithms tested by the authors of the selected studies and identified the ones that achieved the best prediction accuracy in each healthcare scenario. While no algorithm performed consistently, Random Forest performed the best in a few studies. This could serve as a good starting point for future studies on collaborative machine learning on IoMT data.
Coronavirus can be transmitted through the air by close proximity to infected persons. Commercial aircraft are a likely way to both transmit the virus among passengers and move the virus between locations. The importance of learning about where and how coronavirus has entered the United States will help further our understanding of the disease. Air travelers can come from countries or areas with a high rate of infection and may very well be at risk of being exposed to the virus. Therefore, as they reach the United States, the virus could easily spread. On our analysis, we utilized machine learning to determine if the number of flights into the Washington DC Metro Area had an effect on the number of cases and deaths reported in the city and surrounding area.
Recent works on machine learning for combinatorial optimization have shown that learning based approaches can outperform heuristic methods in terms of speed and performance. In this paper, we consider the problem of finding an optimal topological order on a directed acyclic graph with focus on the memory minimization problem which arises in compilers. We propose an end-to-end machine learning based approach for topological ordering using an encoder-decoder framework. Our encoder is a novel attention based graph neural network architecture called \emph{Topoformer} which uses different topological transforms of a DAG for message passing. The node embeddings produced by the encoder are converted into node priorities which are used by the decoder to generate a probability distribution over topological orders. We train our model on a dataset of synthetically generated graphs called layered graphs. We show that our model outperforms, or is on-par, with several topological ordering baselines while being significantly faster on synthetic graphs with up to 2k nodes. We also train and test our model on a set of real-world computation graphs, showing performance improvements.
Machine learning is considered to be one of the most promising applications of quantum computing. Therefore, the search for quantum advantage of the quantum analogues of machine learning models is a key research goal. Here, we show that variational quantum classifiers and support vector machines with quantum kernels can solve a classification problem based on the $k$-Forrelation problem, which is known to be PromiseBQP-complete. Because the PromiseBQP complexity class includes all Bounded-Error Quantum Polynomial-Time (BQP) decision problems, our results imply that there exists a feature map and a quantum kernel that make variational quantum classifiers and quantum kernel support vector machines efficient solvers for any BQP problem. Hence, this work implies that their feature map and quantum kernel, respectively, can be designed to have a quantum advantage for any classification problem that cannot be classically solved in polynomial time but contrariwise by a quantum computer.