A Transfer Learning Evaluation of Deep Neural Networks for Image Classification

Transfer learning is a machine learning technique that uses previously acquired knowledge from a source domain to enhance learning in a target domain by reusing learned weights. This technique is ubiquitous because of its great advantages in achieving high performance while saving training time, memory, and effort in network design. In this paper, we investigate how to select the best pre-trained model that meets the target domain requirements for image classification tasks. In our study, we refined the output layers and general network parameters to apply the knowledge of eleven image processing models, pre-trained on ImageNet, to five different target domain datasets. We measured the accuracy, accuracy density, training time, and model size to evaluate the pre-trained models both in training sessions in one episode and with ten episodes.

A Comparison of Surrogate Modeling Techniques for Global Sensitivity Analysis in Hybrid Simulation

Hybrid simulation is a method used to investigate the dynamic response of a system subjected to a realistic loading scenario. The system under consideration is divided into multiple individual substructures, out of which one or more are tested physically, whereas the remaining are simulated numerically. The coupling of all substructures forms the so-called hybrid model. Although hybrid simulation is extensively used across various engineering disciplines, it is often the case that the hybrid model and related excitation are conceived as being deterministic. However, associated uncertainties are present, whilst simulation deviation, due to their presence, could be significant. In this regard, global sensitivity analysis based on Sobol’ indices can be used to determine the sensitivity of the hybrid model response due to the presence of the associated uncertainties. Nonetheless, estimation of the Sobol’ sensitivity indices requires an unaffordable amount of hybrid simulation evaluations. Therefore, surrogate modeling techniques using machine learning data-driven regression are utilized to alleviate this burden. This study extends the current global sensitivity analysis practices in hybrid simulation by employing various different surrogate modeling methodologies as well as providing comparative results. In particular, polynomial chaos expansion, Kriging and polynomial chaos Kriging are used. A case study encompassing a virtual hybrid model is employed, and hybrid model response quantities of interest are selected. Their respective surrogates are developed, using all three aforementioned techniques. The Sobol’ indices obtained utilizing each examined surrogate are compared with each other, and the results highlight potential deviations when different surrogates are used.

Automated Event Detection and Classification in Soccer: The Potential of Using Multiple Modalities

Detecting events in videos is a complex task, and many different approaches, aimed at a large variety of use-cases, have been proposed in the literature. Most approaches, however, are unimodal and only consider the visual information in the videos. This paper presents and evaluates different approaches based on neural networks where we combine visual features with audio features to detect (spot) and classify events in soccer videos. We employ model fusion to combine different modalities such as video and audio, and test these combinations against different state-of-the-art models on the SoccerNet dataset. The results show that a multimodal approach is beneficial. We also analyze how the tolerance for delays in classification and spotting time, and the tolerance for prediction accuracy, influence the results. Our experiments show that using multiple modalities improves event detection performance for certain types of events.

Detection and Classification of Knee Injuries from MR Images Using the MRNet Dataset with Progressively Operating Deep Learning Methods

This study aimed to build progressively operating deep learning models that could detect meniscus injuries, anterior cruciate ligament (ACL) tears and knee abnormalities in magnetic resonance imaging (MRI). The Stanford Machine Learning Group MRNet dataset was employed in the study, which included MRI image indexes in the coronal, sagittal, and axial axes, each having 1130 trains and 120 validation items. The study is divided into three sections. In the first section, suitable images are selected to determine the disease in the image index based on the disturbance under examination. It is also used to identify images that have been misclassified or are noisy and/or damaged to the degree that they cannot be utilised for diagnosis in the first section. The study employed the 50-layer residual networks (ResNet50) model in this section. The second part of the study involves locating the region to be focused on based on the disturbance that is targeted to be diagnosed in the image under examination. A novel model was built by integrating the convolutional neural networks (CNN) and the denoising autoencoder models in the second section. The third section is dedicated to making a diagnosis of the disease. In this section, a novel ResNet50 model is trained to identify disease diagnoses or abnormalities, independent of the ResNet50 model used in the first section. The images that each model selects as output after training are referred to as progressively operating deep learning methods since they are supplied as an input to the following model.

Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Survey

Deep Learning is a state-of-the-art technique to make inference on extensive or complex data. As a black box model due to their multilayer nonlinear structure, Deep Neural Networks are often criticized as being non-transparent and their predictions not traceable by humans. Furthermore, the models learn from artificially generated datasets, which often do not reflect reality. By basing decision-making algorithms on Deep Neural Networks, prejudice and unfairness may be promoted unknowingly due to a lack of transparency. Hence, several so-called explanators, or explainers, have been developed. Explainers try to give insight into the inner structure of machine learning black boxes by analyzing the connection between the input and output. In this survey, we present the mechanisms and properties of explaining systems for Deep Neural Networks for Computer Vision tasks. We give a comprehensive overview about the taxonomy of related studies and compare several survey papers that deal with explainability in general. We work out the drawbacks and gaps and summarize further research ideas.

AI-Based Video Clipping of Soccer Events

The current gold standard for extracting highlight clips from soccer games is the use of manual annotations and clippings, where human operators define the start and end of an event and trim away the unwanted scenes. This is a tedious, time-consuming, and expensive task, to the extent of being rendered infeasible for use in lower league games. In this paper, we aim to automate the process of highlight generation using logo transition detection, scene boundary detection, and optional scene removal. We experiment with various approaches, using different neural network architectures on different datasets, and present two models that automatically find the appropriate time interval for extracting goal events. These models are evaluated both quantitatively and qualitatively, and the results show that we can detect logo and scene transitions with high accuracy and generate highlight clips that are highly acceptable for viewers. We conclude that there is considerable potential in automating the overall soccer video clipping process.

A Novel Feature Representation for Prediction of Global Horizontal Irradiance Using a Bidirectional Model

Complex weather conditions—in particular clouds—leads to uncertainty in photovoltaic (PV) systems, which makes solar energy prediction very difficult. Currently, in the renewable energy domain, deep-learning-based sequence models have reported better results compared to state-of-the-art machine-learning models. There are quite a few choices of deep-learning architectures, among which Bidirectional Gated Recurrent Unit (BGRU) has apparently not been used earlier in the solar energy domain. In this paper, BGRU was used with a new augmented and bidirectional feature representation. The used BGRU network is more generalized as it can handle unequal lengths of forward and backward context. The proposed model produced 59.21%, 37.47%, and 76.80% better prediction accuracy compared to traditional sequence-based, bidirectional models, and some of the established states-of-the-art models. The testbed considered for evaluation of the model is far more comprehensive and reliable considering the variability in the climatic zones and seasons, as compared to some of the recent studies in India.

Language Semantics Interpretation with an Interaction-Based Recurrent Neural Network

Text classification is a fundamental language task in Natural Language Processing. A variety of sequential models are capable of making good predictions, yet there is a lack of connection between language semantics and prediction results. This paper proposes a novel influence score (I-score), a greedy search algorithm, called Backward Dropping Algorithm (BDA), and a novel feature engineering technique called the “dagger technique”. First, the paper proposes to use the novel influence score (I-score) to detect and search for the important language semantics in text documents that are useful for making good predictions in text classification tasks. Next, a greedy search algorithm, called the Backward Dropping Algorithm, is proposed to handle long-term dependencies in the dataset. Moreover, the paper proposes a novel engineering technique called the “dagger technique” that fully preserves the relationship between the explanatory variable and the response variable. The proposed techniques can be further generalized into any feed-forward Artificial Neural Networks (ANNs) and Convolutional Neural Networks (CNNs), and any neural network. A real-world application on the Internet Movie Database (IMDB) is used and the proposed methods are applied to improve prediction performance with an 81% error reduction compared to other popular peers if I-score and “dagger technique” are not implemented.

A Multi-Component Framework for the Analysis and Design of Explainable Artificial Intelligence

The rapid growth of research in explainable artificial intelligence (XAI) follows on two substantial developments. First, the enormous application success of modern machine learning methods, especially deep and reinforcement learning, have created high expectations for industrial, commercial, and social value. Second, the emerging and growing concern for creating ethical and trusted AI systems, including compliance with regulatory principles to ensure transparency and trust. These two threads have created a kind of “perfect storm” of research activity, all motivated to create and deliver any set of tools and techniques to address the XAI demand. As some surveys of current XAI suggest, there is yet to appear a principled framework that respects the literature of explainability in the history of science and which provides a basis for the development of a framework for transparent XAI. We identify four foundational components, including the requirements for (1) explicit explanation knowledge representation, (2) delivery of alternative explanations, (3) adjusting explanations based on knowledge of the explainee, and (4) exploiting the advantage of interactive explanation. With those four components in mind, we intend to provide a strategic inventory of XAI requirements, demonstrate their connection to a basic history of XAI ideas, and then synthesize those ideas into a simple framework that can guide the design of AI systems that require XAI.

Deep Self-Organizing Map of Convolutional Layers for Clustering and Visualizing Image Data

The self-organizing convolutional map (SOCOM) hybridizes convolutional neural networks, self-organizing maps, and gradient backpropagation optimization into a novel integrated unsupervised deep learning model. SOCOM structurally combines, architecturally stacks, and algorithmically fuses its deep/unsupervised learning components. The higher-level representations produced by its underlying convolutional deep architecture are embedded in its topologically ordered neural map output. The ensuing unsupervised clustering and visualization operations reflect the model’s degree of synergy between its building blocks and synopsize its range of applications. Clustering results are reported on the STL-10 benchmark dataset coupled with the devised neural map visualizations. The series of conducted experiments utilize a deep VGG-based SOCOM model.