When COVID-19 first started spreading and quarantine was implemented, the Society for Industrial and Applied Mathematics (SIAM) Student Chapter at the University of Minnesota-Twin Cities began a collaboration with Ecolab to use our skills as data scientists and mathematicians to extract useful insights from relevant data relating to the pandemic. This collaboration consisted of multiple groups working on different projects. In this write-up we focus on using clustering techniques to help us find groups of similar counties in the US and use that to help us understand the pandemic. Our team for this project consisted of University of Minnesota students Cora Brown, Sarah Milstein, Tianyi Sun, and Cooper Zhao, with help from Ecolab Data Scientist Jimmy Broomfield and University of Minnesota student Skye Ke. In the sections below we describe all of the work done for this project. In Section 2, we list the data we gathered, as well as the feature engineering we performed. In Section 3, we describe the metrics we used for evaluating our models. In Section 4, we explain the methods we used for interpreting the results of our various clustering approaches. In Section 5, we describe the different clustering methods we implemented. In Section 6, we present the results of our clustering techniques and provide relevant interpretation. Finally, in Section 7, we provide some concluding remarks comparing the different clustering methods.
The continuously growing number of objects orbiting around the Earth is expected to be accompanied by an increasing frequency of objects re-entering the Earth's atmosphere. Many of these re-entries will be uncontrolled, making their prediction challenging and subject to several uncertainties. Traditionally, re-entry predictions are based on the propagation of the object's dynamics using state-of-the-art modelling techniques for the forces acting on the object. However, modelling errors, particularly related to the prediction of atmospheric drag may result in poor prediction accuracies. In this context, we explore the possibility to perform a paradigm shift, from a physics-based approach to a data-driven approach. To this aim, we present the development of a deep learning model for the re-entry prediction of uncontrolled objects in Low Earth Orbit (LEO). The model is based on a modified version of the Sequence-to-Sequence architecture and is trained on the average altitude profile as derived from a set of Two-Line Element (TLE) data of over 400 bodies. The novelty of the work consists in introducing in the deep learning model, alongside the average altitude, three new input features: a drag-like coefficient (B*), the average solar index, and the area-to-mass ratio of the object. The developed model is tested on a set of objects studied in the Inter-Agency Space Debris Coordination Committee (IADC) campaigns. The results show that the best performances are obtained on bodies characterised by the same drag-like coefficient and eccentricity distribution as the training set.
On a daily basis, data centers process huge volumes of data backed by the proliferation of inexpensive hard disks. Data stored in these disks serve a range of critical functional needs from financial, and healthcare to aerospace. As such, premature disk failure and consequent loss of data can be catastrophic. To mitigate the risk of failures, cloud storage providers perform condition-based monitoring and replace hard disks before they fail. By estimating the remaining useful life of hard disk drives, one can predict the time-to-failure of a particular device and replace it at the right time, ensuring maximum utilization whilst reducing operational costs. In this work, large-scale predictive analyses are performed using severely skewed health statistics data by incorporating customized feature engineering and a suite of sequence learners. Past work suggests using LSTMs as an excellent approach to predicting remaining useful life. To this end, we present an encoder-decoder LSTM model where the context gained from understanding health statistics sequences aid in predicting an output sequence of the number of days remaining before a disk potentially fails. The models developed in this work are trained and tested across an exhaustive set of all of the 10 years of S.M.A.R.T. health data in circulation from Backblaze and on a wide variety of disk instances. It closes the knowledge gap on what full-scale training achieves on thousands of devices and advances the state-of-the-art by providing tangible metrics for evaluation and generalization for practitioners looking to extend their workflow to all years of health data in circulation across disk manufacturers. The encoder-decoder LSTM posted an RMSE of 0.83 during training and 0.86 during testing over the exhaustive 10 year data while being able to generalize competitively over other drives from the Seagate family.
Iris recognition is a secure biometric technology known for its stability and privacy. With no two irises being identical and little change throughout a person's lifetime, iris recognition is considered more reliable and less susceptible to external factors than other biometric recognition methods. Unlike traditional machine learning-based iris recognition methods, deep learning technology does not rely on feature engineering and boasts excellent performance. This paper collects 120 relevant papers to summarize the development of iris recognition based on deep learning. We first introduce the background of iris recognition and the motivation and contribution of this survey. Then, we present the common datasets widely used in iris recognition. After that, we summarize the key tasks involved in the process of iris recognition based on deep learning technology, including identification, segmentation, presentation attack detection, and localization. Finally, we discuss the challenges and potential development of iris recognition. This review provides a comprehensive sight of the research of iris recognition based on deep learning.
The performance of fault diagnosis systems is highly affected by data quality in cyber-physical power systems. These systems generate massive amounts of data that overburden the system with excessive computational costs. Another issue is the presence of noise in recorded measurements, which prevents building a precise decision model. Furthermore, the diagnostic model is often provided with a mixture of redundant measurements that may deviate it from learning normal and fault distributions. This paper presents the effect of feature engineering on mitigating the aforementioned challenges in cyber-physical systems. Feature selection and dimensionality reduction methods are combined with decision models to simulate data-driven fault diagnosis in a 118-bus power system. A comparative study is enabled accordingly to compare several advanced techniques in both domains. Dimensionality reduction and feature selection methods are compared both jointly and separately. Finally, experiments are concluded, and a setting is suggested that enhances data quality for fault diagnosis.
We present a robust deep incremental learning framework for regression tasks on financial temporal tabular datasets which is built upon the incremental use of commonly available tabular and time series prediction models to adapt to distributional shifts typical of financial datasets. The framework uses a simple basic building block (decision trees) to build self-similar models of any required complexity to deliver robust performance under adverse situations such as regime changes, fat-tailed distributions, and low signal-to-noise ratios. As a detailed study, we demonstrate our scheme using XGBoost models trained on the Numerai dataset and show that a two layer deep ensemble of XGBoost models over different model snapshots delivers high quality predictions under different market regimes. We also show that the performance of XGBoost models with different number of boosting rounds in three scenarios (small, standard and large) is monotonically increasing with respect to model size and converges towards the generalisation upper bound. We also evaluate the robustness of the model under variability of different hyperparameters, such as model complexity and data sampling settings. Our model has low hardware requirements as no specialised neural architectures are used and each base model can be independently trained in parallel.
Deep learning (DL) algorithms have been widely applied to short-term voltage stability (STVS) assessment in power systems. However, transferring the knowledge learned in one power grid to other power grids with topology changes is still a challenging task. This paper proposed a transferable DL-based model for STVS assessment by constructing the topology-aware voltage dynamic features from raw PMU data. Since the reactive power flow and grid topology are essential to voltage stability, the topology-aware and physics-informed voltage dynamic features are utilized to effectively represent the topological and temporal patterns from post-disturbance system dynamic trajectories. The proposed DL-based STVS assessment model is tested under random operating conditions on the New England 39-bus system. It has 99.99\% classification accuracy of the short-term voltage stability status using the topology-aware and physics-informed voltage dynamic features. In addition to high accuracy, the experiments show good adaptability to PMU errors. Moreover, The proposed STVS assessment method has outstanding performance on new grid topologies after fine-tuning. In particular, the highest accuracy reaches 99.68\% in evaluation, which demonstrates a good knowledge transfer ability of the proposed model for power grid topology change.
Although deep neural models substantially reduce the overhead of feature engineering, the features readily available in the inputs might significantly impact training cost and the performance of the models. In this paper, we explore the impact of an unsuperivsed feature enrichment approach based on variable roles on the performance of neural models of code. The notion of variable roles (as introduced in the works of Sajaniemi et al. [Refs. 1,2]) has been found to help students' abilities in programming. In this paper, we investigate if this notion would improve the performance of neural models of code. To the best of our knowledge, this is the first work to investigate how Sajaniemi et al.'s concept of variable roles can affect neural models of code. In particular, we enrich a source code dataset by adding the role of individual variables in the dataset programs, and thereby conduct a study on the impact of variable role enrichment in training the Code2Seq model. In addition, we shed light on some challenges and opportunities in feature enrichment for neural code intelligence models.
Synthesizing controllable motion for a character using deep learning has been a promising approach due to its potential to learn a compact model without laborious feature engineering. To produce dynamic motion from weak control signals such as desired paths, existing methods often require auxiliary information such as phases for alleviating motion ambiguity, which limits their generalisation capability. As past poses often contain useful auxiliary hints, in this paper, we propose a task-agnostic deep learning method, namely Multi-scale Control Signal-aware Transformer (MCS-T), with an attention based encoder-decoder architecture to discover the auxiliary information implicitly for synthesizing controllable motion without explicitly requiring auxiliary information such as phase. Specifically, an encoder is devised to adaptively formulate the motion patterns of a character's past poses with multi-scale skeletons, and a decoder driven by control signals to further synthesize and predict the character's state by paying context-specialised attention to the encoded past motion patterns. As a result, it helps alleviate the issues of low responsiveness and slow transition which often happen in conventional methods not using auxiliary information. Both qualitative and quantitative experimental results on an existing biped locomotion dataset, which involves diverse types of motion transitions, demonstrate the effectiveness of our method. In particular, MCS-T is able to successfully generate motions comparable to those generated by the methods using auxiliary information.
As cellular networks evolve towards the 6th generation, machine learning is seen as a key enabling technology to improve the capabilities of the network. Machine learning provides a methodology for predictive systems, which can make networks become proactive. This proactive behavior of the network can be leveraged to sustain, for example, a specific quality of service requirement. With predictive quality of service, a wide variety of new use cases, both safety- and entertainment-related, are emerging, especially in the automotive sector. Therefore, in this work, we consider maximum throughput prediction enhancing, for example, streaming or high-definition mapping applications. We discuss the entire machine learning workflow highlighting less regarded aspects such as the detailed sampling procedures, the in-depth analysis of the dataset characteristics, the effects of splits in the provided results, and the data availability. Reliable machine learning models need to face a lot of challenges during their lifecycle. We highlight how confidence can be built on machine learning technologies by better understanding the underlying characteristics of the collected data. We discuss feature engineering and the effects of different splits for the training processes, showcasing that random splits might overestimate performance by more than twofold. Moreover, we investigate diverse sets of input features, where network information proved to be most effective, cutting the error by half. Part of our contribution is the validation of multiple machine learning models within diverse scenarios. We also use explainable AI to show that machine learning can learn underlying principles of wireless networks without being explicitly programmed. Our data is collected from a deployed network that was under full control of the measurement team and covered different vehicular scenarios and radio environments.
Diffractive optical neural networks (DONNs) have been emerging as a high-throughput and energy-efficient hardware platform to perform all-optical machine learning (ML) in machine vision systems. However, the current demonstrated applications of DONNs are largely straightforward image classification tasks, which undermines the prospect of developing and utilizing such hardware for other ML applications. Here, we numerically and experimentally demonstrate the deployment of an all-optical reconfigurable DONNs system for scientific computing, including guiding two-dimensional quantum material synthesis, predicting the properties of nanomaterials and small molecular cancer drugs, predicting the device response of nanopatterned integrated photonic power splitters, and the dynamic stabilization of an inverted pendulum with reinforcement learning. Despite a large variety of input data structures, we develop a universal feature engineering approach to convert categorical input features to the images that can be processed in the DONNs system. Our results open up new opportunities of employing DONNs systems for a broad range of ML applications.
Businesses and customers can gain valuable information from product reviews. The sheer number of reviews often necessitates ranking them based on their potential helpfulness. However, only a few reviews ever receive any helpfulness votes on online marketplaces. Sorting all reviews based on the few existing votes can cause helpful reviews to go unnoticed because of the limited attention span of readers. The problem of review helpfulness prediction is even more important for higher review volumes, and newly written reviews or launched products. In this work we compare the use of RoBERTa and XLM-R language models to predict the helpfulness of online product reviews. The contributions of our work in relation to literature include extensively investigating the efficacy of state-of-the-art language models -- both monolingual and multilingual -- against a robust baseline, taking ranking metrics into account when assessing these approaches, and assessing multilingual models for the first time. We employ the Amazon review dataset for our experiments. According to our study on several product categories, multilingual and monolingual pre-trained language models outperform the baseline that utilizes random forest with handcrafted features as much as 23% in RMSE. Pre-trained language models reduce the need for complex text feature engineering. However, our results suggest that pre-trained multilingual models may not be used for fine-tuning only one language. We assess the performance of language models with and without additional features. Our results show that including additional features like product rating by the reviewer can further help the predictive methods.
Real-world degradations deviate from ideal degradations, as most deep learning-based scenarios involve the ideal synthesis of low-resolution (LR) counterpart images by popularly used bicubic interpolation. Moreover, supervised learning approaches rely on many high-resolution (HR) and LR image pairings to reconstruct missing information based on their association, developed by complex long hours of deep neural network training. Additionally, the trained model's generalizability on various image datasets with various distributions is not guaranteed. To overcome this challenge, we proposed our novel Self-FuseNet, particularly for extremely poor-resolution satellite images. Also, the network exhibits strong generalization performance on additional datasets (both “ideal” and “nonideal” scenarios). The network is especially for those image datasets suffering from the following two significant limitations: 1) nonavailability of ground truth HR images; 2) limitation of a large count of the unpaired dataset for deep neural network training. The benefit of the proposed model is threefold: 1) it does not require any significant extensive training data, either paired or unpaired but only a single LR image without prior knowledge of its distribution; 2) it is a simple and effective model for super-resolving very poor-resolution images, saving computational resources and time; 3) using UNet, the processing of data are accelerated by the network's wide skip connections, allowing image reconstruction with fewer parameters. Rather than using an inverse approach, as common in most deep learning scenarios, we introduced a forward approach to super-resolve exceptionally LR remote sensing images. This demonstrates its supremacy over recently proposed state-of-the-art methods for unsupervised single real-world image blind super-resolution.
Several explainable AI methods allow a Machine Learning user to get insights on the classification process of a black-box model in the form of local linear explanations. With such information, the user can judge which features are locally relevant for the classification outcome, and get an understanding of how the model reasons. Standard supervised learning processes are purely driven by the original features and target labels, without any feedback loop informed by the local relevance of the features identified by the post-hoc explanations. In this paper, we exploit this newly obtained information to design a feature engineering phase, where we combine explanations with feature values. To do so, we develop two different strategies, named Iterative Dataset Weighting and Targeted Replacement Values, which generate streamlined models that better mimic the explanation process presented to the user. We show how these streamlined models compare to the original black-box classifiers, in terms of accuracy and compactness of the newly produced explanations.
Blazars are active galactic nuclei with relativistic jets pointed almost directly at Earth. Blazars are characterized by strong, apparently stochastic flux variability at virtually all observed wavelengths and timescales, from minutes to years, the physical origin of which is still poorly understood. In the high-energy gamma-ray band, the Large Area Telescope aboard the Fermi space telescope (Fermi-LAT) has conducted regular monitoring of thousands of blazars since 2008. Deep learning can help uncover structure in gamma-ray blazars' complex variability patterns that traditional methods based on parametric statistical modeling or manual feature engineering may miss. In this work, we propose using a self-supervised Transformer encoder architecture to construct an effective representation of blazar gamma-ray variability. Measurement errors, upper limits, and missing data are accommodated using learned encodings. The model predicts a set of quantiles for the flux probability distribution at each time step, an architecture naturally suited for describing data generated by a stochastic process. As a proof of concept for how the model output can be analyzed to extract scientifically relevant information, a preliminary search for weekly-timescale time-reversal asymmetry in gamma-ray blazar light curves was conducted, finding no significant evidence for asymmetry.
Retrieving the similar solutions from the historical case base for new design requirements is the first step in mechanical part redesign under the context of case-based reasoning. However, the manual retrieving method has the problem of low efficiency when the case base is large. Additionally, it is difficult for simple reasoning algorithms (e.g., rule-based reasoning, decision tree) to cover all the features in complicated design solutions. In this regard, a text2shape deep retrieval model is established in order to support text description-based mechanical part shapes retrieval, where the texts are for describing the structural features of the target mechanical parts. More specifically, feature engineering is applied to identify the key structural features of the target mechanical parts. Based on the identified key structural features, a training set of 1000 samples was constructed, where each sample consisted of a paragraph of text description of a group of structural features and the corresponding 3D shape of the structural features. RNN and 3D CNN algorithms were customized to build the text2shape deep retrieval model. Orthogonal experiments were used for modeling turning. Eventually, the highest accuracy of the model was 0.98; therefore, the model can be effective for retrieving initial cases for mechanical part redesign.
As more and more Arabic texts emerged on the Internet, extracting important information from these Arabic texts is especially useful. As a fundamental technology, Named entity recognition (NER) serves as the core component in information extraction technology, while also playing a critical role in many other Natural Language Processing (NLP) systems, such as question answering and knowledge graph building. In this paper, we provide a comprehensive review of the development of Arabic NER, especially the recent advances in deep learning and pre-trained language model. Specifically, we first introduce the background of Arabic NER, including the characteristics of Arabic and existing resources for Arabic NER. Then, we systematically review the development of Arabic NER methods. Traditional Arabic NER systems focus on feature engineering and designing domain-specific rules. In recent years, deep learning methods achieve significant progress by representing texts via continuous vector representations. With the growth of pre-trained language model, Arabic NER yields better performance. Finally, we conclude the method gap between Arabic NER and NER methods from other languages, which helps outline future directions for Arabic NER.
In recent years, the employment of deep learning methods has led to several significant breakthroughs in artificial intelligence. Different from traditional machine learning models, deep learning-based approaches are able to extract features autonomously from raw data. This allows for bypassing the feature engineering process, which is generally considered to be both error-prone and tedious. Moreover, deep learning strategies often outperform traditional models in terms of accuracy.
There exist growing interests in intelligent systems for numerous medical imaging, image processing, and computer vision applications, such as face recognition, medical diagnosis, character recognition, and self-driving cars, among others. These applications usually require solving complex classification problems involving complex images with unknown data generative processes. In addition to recent successes of the current classification approaches relying on feature engineering and deep learning, several shortcomings of them, such as the lack of robustness, generalizability, and interpretability, have also been observed. These methods often require extensive training data, are computationally expensive, and are vulnerable to out-of-distribution samples, e.g., adversarial attacks. Recently, an accurate, data-efficient, computationally efficient, and robust transport-based classification approach has been proposed, which describes a generative model-based problem formulation and closed-form solution for a specific category of classification problems. However, all these approaches lack mechanisms to detect test samples outside the class distributions used during training. In real-world settings, where the collected training samples are unable to exhaust or cover all classes, the traditional classification schemes are unable to handle the unseen classes effectively, which is especially an important issue for safety-critical systems, such as self-driving and medical imaging diagnosis. In this work, we propose a method for detecting out-of-class distributions based on the distribution of sliced-Wasserstein distance from the Radon Cumulative Distribution Transform (R-CDT) subspace. We tested our method on the MNIST and two medical image datasets and reported better accuracy than the state-of-the-art methods without an out-of-class distribution detection procedure.
In this work, we propose a new class of learnable optimizers, called \textit{Mnemosyne}. It is based on the novel spatio-temporal low-rank implicit attention Transformers that can learn to train entire neural network architectures, including other Transformers, without any task-specific optimizer tuning. We show that Mnemosyne: (a) outperforms popular LSTM optimizers (also with new feature engineering to mitigate catastrophic forgetting of LSTMs), (b) can successfully train Transformers while using simple meta-training strategies that require minimal computational resources, (c) matches accuracy-wise SOTA hand-designed optimizers with carefully tuned hyper-parameters (often producing top performing models). Furthermore, Mnemosyne provides space complexity comparable to that of its hand-designed first-order counterparts, which allows it to scale to training larger sets of parameters. We conduct an extensive empirical evaluation of Mnemosyne on: (a) fine-tuning a wide range of Vision Transformers (ViTs) from medium-size architectures to massive ViT-Hs (36 layers, 16 heads), (b) pre-training BERT models and (c) soft prompt-tuning large 11B+ T5XXL models. We complement our results with a comprehensive theoretical analysis of the compact associative memory used by Mnemosyne which we believe was never done before.