In wireless communications, estimation of channels in OFDM systems spans frequency and time, which relies on sparse collections of pilot data, posing an ill-posed inverse problem. Moreover, deep learning estimators require large amounts of training data, computational resources, and true channels to produce accurate channel estimates, which are not realistic. To address this, a convolutional neural tangent kernel (CNTK) is derived from an infinitely wide convolutional network whose training dynamics can be expressed by a closed-form equation. This CNTK is used to impute the target matrix and estimate the missing channel response using only the known values available at pilot locations. This is a promising solution for channel estimation that does not require a large training set. Numerical results on realistic channel datasets demonstrate that our strategy accurately estimates the channels without a large dataset and significantly outperforms deep learning methods in terms of speed, accuracy, and computational resources.
Deep learning-based electrocardiogram (ECG) classification has shown impressive performance but clinical adoption has been slowed by the lack of transparent and faithful explanations. Post hoc methods such as saliency maps may fail to reflect a model's true decision process. Prototype-based reasoning offers a more transparent alternative by grounding decisions in similarity to learned representations of real ECG segments, enabling faithful, case-based explanations. We introduce ProtoECGNet, a prototype-based deep learning model for interpretable, multi-label ECG classification. ProtoECGNet employs a structured, multi-branch architecture that reflects clinical interpretation workflows: it integrates a 1D CNN with global prototypes for rhythm classification, a 2D CNN with time-localized prototypes for morphology-based reasoning, and a 2D CNN with global prototypes for diffuse abnormalities. Each branch is trained with a prototype loss designed for multi-label learning, combining clustering, separation, diversity, and a novel contrastive loss that encourages appropriate separation between prototypes of unrelated classes while allowing clustering for frequently co-occurring diagnoses. We evaluate ProtoECGNet on all 71 diagnostic labels from the PTB-XL dataset, demonstrating competitive performance relative to state-of-the-art black-box models while providing structured, case-based explanations. To assess prototype quality, we conduct a structured clinician review of the final model's projected prototypes, finding that they are rated as representative and clear. ProtoECGNet shows that prototype learning can be effectively scaled to complex, multi-label time-series classification, offering a practical path toward transparent and trustworthy deep learning models for clinical decision support.
One of the key safety considerations of battery manufacturing is thermal runaway, the uncontrolled increase in temperature which can lead to fires, explosions, and emissions of toxic gasses. As such, development of automated systems capable of detecting such events is of considerable importance in both academic and industrial contexts. In this work, we investigate the use of deep learning for detecting thermal runaway in the battery production line of VDL Nedcar, a Dutch automobile manufacturer. Specifically, we collect data from the production line to represent both baseline (non thermal runaway) and thermal runaway conditions. Thermal runaway was simulated through the use of external heat and smoke sources. The data consisted of both optical and thermal images which were then preprocessed and fused before serving as input to our models. In this regard, we evaluated three deep-learning models widely used in computer vision including shallow convolutional neural networks, residual neural networks, and vision transformers on two performance metrics. Furthermore, we evaluated these models using explainability methods to gain insight into their ability to capture the relevant feature information from their inputs. The obtained results indicate that the use of deep learning is a viable approach to thermal runaway detection in battery production lines.
We present a theoretically well-founded deep learning algorithm for nonparametric regression. It uses over-parametrized deep neural networks with logistic activation function, which are fitted to the given data via gradient descent. We propose a special topology of these networks, a special random initialization of the weights, and a data-dependent choice of the learning rate and the number of gradient descent steps. We prove a theoretical bound on the expected $L_2$ error of this estimate, and illustrate its finite sample size performance by applying it to simulated data. Our results show that a theoretical analysis of deep learning which takes into account simultaneously optimization, generalization and approximation can result in a new deep learning estimate which has an improved finite sample performance.
Automatic Modulation Classification (AMC) is a signal processing technique widely used at the physical layer of wireless systems to enhance spectrum utilization efficiency. In this work, we propose a fast and accurate AMC system, termed DL-AMC, which leverages deep learning techniques. Specifically, DL-AMC is built using convolutional neural network (CNN) architectures, including ResNet-18, ResNet-50, and MobileNetv2. To evaluate its performance, we curated a comprehensive dataset containing various modulation schemes. Each modulation type was transformed into an eye diagram, with signal-to-noise ratio (SNR) values ranging from -20 dB to 30 dB. We trained the CNN models on this dataset to enable them to learn the discriminative features of each modulation class effectively. Experimental results show that the proposed DL-AMC models achieve high classification accuracy, especially in low SNR conditions. These results highlight the robustness and efficacy of DL-AMC in accurately classifying modulations in challenging wireless environments
Machine learning research in GNSS radio frequency interference (RFI) detection often lacks a clear empirical justification for the choice of deep learning architectures over simpler machine learning approaches. In this work, we argue for a change in research direction-from developing ever more complex deep learning models to carefully assessing their real-world effectiveness in comparison to interpretable and lightweight machine learning baselines. Our findings reveal that state-of-the-art deep learning models frequently fail to outperform simple, well-engineered machine learning methods in the context of GNSS RFI detection. Leveraging a unique large-scale dataset collected by the Swiss Air Force and Swiss Air-Rescue (Rega), and preprocessed by Swiss Air Navigation Services Ltd. (Skyguide), we demonstrate that a simple baseline model achieves 91\% accuracy in detecting GNSS RFI, outperforming more complex deep learning counterparts. These results highlight the effectiveness of pragmatic solutions and offer valuable insights to guide future research in this critical application domain.
Loss functions are at the heart of deep learning, shaping how models learn and perform across diverse tasks. They are used to quantify the difference between predicted outputs and ground truth labels, guiding the optimization process to minimize errors. Selecting the right loss function is critical, as it directly impacts model convergence, generalization, and overall performance across various applications, from computer vision to time series forecasting. This paper presents a comprehensive review of loss functions, covering fundamental metrics like Mean Squared Error and Cross-Entropy to advanced functions such as Adversarial and Diffusion losses. We explore their mathematical foundations, impact on model training, and strategic selection for various applications, including computer vision (Discriminative and generative), tabular data prediction, and time series forecasting. For each of these categories, we discuss the most used loss functions in the recent advancements of deep learning techniques. Also, this review explore the historical evolution, computational efficiency, and ongoing challenges in loss function design, underlining the need for more adaptive and robust solutions. Emphasis is placed on complex scenarios involving multi-modal data, class imbalances, and real-world constraints. Finally, we identify key future directions, advocating for loss functions that enhance interpretability, scalability, and generalization, leading to more effective and resilient deep learning models.
This work aimed to present a bibliometric analysis of deep learning research for plant disease identification, with a special focus on generative modeling. A thorough analysis of SCOPUS-sourced bibliometric data from 253 documents was performed. Key performance metrics such as accuracy, precision, recall, and F1-score were analyzed for generative modeling. The findings highlighted significant contributions from some authors Too and Arnal Barbedo, whose works had notable citation counts, suggesting their influence on the academic community. Co-authorship networks revealed strong collaborative clusters, while keyword analysis identified emerging research gaps. This study highlights the role of collaboration and citation metrics in shaping research directions and enhancing the impact of scholarly work in applications of deep learning to plant disease identification. Future research should explore the methodologies of highly cited studies to inform best practices and policy-making.
Learning about underlying patterns in data using latent unobserved structures to improve the accuracy of predictive models has become an active avenue of deep learning research. Most approaches cluster the original features to capture certain latent structures. However, the information gained in the process can often be implicitly derived by sufficiently complex models. Thus, such approaches often provide minimal benefits. We propose a SHAP (Shapley Additive exPlanations) based supervised deep learning framework FORCE which relies on two-stage usage of SHAP values in the neural network architecture, (i) an additional latent feature to guide model training, based on clustering SHAP values, and (ii) initiating an attention mechanism within the architecture using latent information. This approach gives a neural network an indication about the effect of unobserved values that modify feature importance for an observation. The proposed framework is evaluated on three real life datasets. Our results demonstrate that FORCE led to dramatic improvements in overall performance as compared to networks that did not incorporate the latent feature and attention framework (e.g., F1 score for presence of heart disease 0.80 vs 0.72). Using cluster assignments and attention based on SHAP values guides deep learning, enhancing latent pattern learning and overall discriminative capability.
The performance of deep learning methods critically depends on the quality and quantity of the available training data. This is especially the case for physiological time series, which are both noisy and scarce, which calls for data augmentation to artificially increase the size of datasets. Another issue is that the time-evolving statistical properties of nonstationary signals prevent the use of standard data augmentation techniques. To this end, we introduce a novel method for augmenting nonstationary time series. This is achieved by combining offline changepoint detection with the iterative amplitude-adjusted Fourier transform (iAAFT), which ensures that the time-frequency properties of the original signal are preserved during augmentation. The proposed method is validated through comparisons of the performance of i) a deep learning seizure detection algorithm on both the original and augmented versions of the CHB-MIT and Siena scalp electroencephalography (EEG) databases, and ii) a deep learning atrial fibrillation (AF) detection algorithm on the original and augmented versions of the Computing in Cardiology Challenge 2017 dataset. By virtue of the proposed method, for the CHB-MIT and Siena datasets respectively, accuracy rose by 4.4% and 1.9%, precision by 10% and 5.5%, recall by 3.6% and 0.9%, and F1 by 4.2% and 1.4%. For the AF classification task, accuracy rose by 0.3%, precision by 2.1%, recall by 0.8%, and F1 by 2.1%.
Mohs micrographic surgery (MMS) is the gold standard technique for removing high risk nonmelanoma skin cancer however, intraoperative histopathological examination demands significant time, effort, and professionality. The objective of this study is to develop a deep learning model to detect basal cell carcinoma (BCC) and artifacts on Mohs slides. A total of 731 Mohs slides from 51 patients with BCCs were used in this study, with 91 containing tumor and 640 without tumor which was defined as non-tumor. The dataset was employed to train U-Net based models that segment tumor and non-tumor regions on the slides. The segmented patches were classified as tumor, or non-tumor to produce predictions for whole slide images (WSIs). For the segmentation phase, the deep learning model success was measured using a Dice score with 0.70 and 0.67 value, area under the curve (AUC) score with 0.98 and 0.96 for tumor and non-tumor, respectively. For the tumor classification, an AUC of 0.98 for patch-based detection, and AUC of 0.91 for slide-based detection was obtained on the test dataset. We present an AI system that can detect tumors and non-tumors in Mohs slides with high success. Deep learning can aid Mohs surgeons and dermatopathologists in making more accurate decisions.
Traditional physics-based models of geophysical flows, such as debris flows and landslides that pose significant risks to human lives and infrastructure are computationally expensive, limiting their utility for large-scale parameter sweeps, uncertainty quantification, inversions or real-time applications. This study presents an efficient alternative, a deep learning-based surrogate model built using a modified U-Net architecture to predict the dynamics of runoff-generated debris flows across diverse terrain based on data from physics based simulations. The study area is divided into smaller patches for localized predictions using a patch-predict-stitch methodology (complemented by limited global data to accelerate training). The patches are then combined to reconstruct spatially continuous flow maps, ensuring scalability for large domains. To enable fast training using limited expensive simulations, the deep learning model was trained on data from an ensemble of physics based simulations using parameters generated via Latin Hypercube Sampling and validated on unseen parameter sets and terrain, achieving maximum pointwise errors below 10% and robust generalization. Uncertainty quantification using Monte Carlo methods are enabled using the validated surrogate, which can facilitate probabilistic hazard assessments. This study highlights the potential of deep learning surrogates as powerful tools for geophysical flow analysis, enabling computationally efficient and reliable probabilistic hazard map predictions.
Auctions are important mechanisms extensively implemented in various markets, e.g., search engines' keyword auctions, antique auctions, etc. Finding an optimal auction mechanism is extremely difficult due to the constraints of imperfect information, incentive compatibility (IC), and individual rationality (IR). In addition to the traditional economic methods, some recently attempted to find the optimal (single) auction using deep learning methods. Unlike those attempts focusing on single auctions, we develop deep learning methods for double auctions, where imperfect information exists on both the demand and supply sides. The previous attempts on single auction cannot directly apply to our contexts and those attempts additionally suffer from limited generalizability, inefficiency in ensuring the constraints, and learning fluctuations. We innovate in designing deep learning models for solving the more complex problem and additionally addressing the previous models' three limitations. Specifically, we achieve generalizability by leveraging a transformer-based architecture to model market participants as sequences for varying market sizes; we utilize the numerical features of the constraints and pre-treat them for a higher learning efficiency; we develop a gradient-conflict-elimination scheme to address the problem of learning fluctuation. Extensive experimental evaluations demonstrate the superiority of our approach to classical and machine learning baselines.
Relational databases, organized into tables connected by primary-foreign key relationships, are a common format for organizing data. Making predictions on relational data often involves transforming them into a flat tabular format through table joins and feature engineering, which serve as input to tabular methods. However, designing features that fully capture complex relational patterns remains challenging. Graph Neural Networks (GNNs) offer a compelling alternative by inherently modeling these relationships, but their time overhead during inference limits their applicability for real-time scenarios. In this work, we aim to bridge this gap by leveraging existing feature engineering efforts to enhance the efficiency of GNNs in relational databases. Specifically, we use GNNs to capture complex relationships within relational databases, patterns that are difficult to featurize, while employing engineered features to encode temporal information, thereby avoiding the need to retain the entire historical graph and enabling the use of smaller, more efficient graphs. Our \textsc{LightRDL} approach not only improves efficiency, but also outperforms existing models. Experimental results on the RelBench benchmark demonstrate that our framework achieves up to $33\%$ performance improvement and a $526\times$ inference speedup compared to GNNs, making it highly suitable for real-time inference.
Intrusion Detection Systems (IDS) have long been a hot topic in the cybersecurity community. In recent years, with the introduction of deep learning (DL) techniques, IDS have made great progress due to their increasing generalizability. The rationale behind this is that by learning the underlying patterns of known system behaviors, IDS detection can be generalized to intrusions that exploit zero-day vulnerabilities. In this survey, we refer to this type of IDS as DL-based IDS (DL-IDS). From the perspective of DL, this survey systematically reviews all the stages of DL-IDS, including data collection, log storage, log parsing, graph summarization, attack detection, and attack investigation. To accommodate current researchers, a section describing the publicly available benchmark datasets is included. This survey further discusses current challenges and potential future research directions, aiming to help researchers understand the basic ideas and visions of DL-IDS research, as well as to motivate their research interests.
Resource-optimized deep neural networks (DNNs) nowadays run on microcontrollers to perform a wide variety of audio, image and sensor data classification tasks. Despite comprehensive support for deep learning tools for 32-bit microcontrollers, performing deep learning inferences on 16-bit microcontrollers still remains a chal-lenge. Although there are some tools for implementing neural net-works on 16-bit systems, generally, there is a large gap in efficiency between the development tools for 16-bit microcontrollers and 32-bit (or higher) systems. There is also a steep learning curve that discourages beginners inexperienced with microcontrollers and programming in C to develop efficient and effective deep learning models for 16-bit microcontrollers. To fill this gap, we have created a neural network model generator that (1) automatically transfers parameters of a pre-trained DNN or CNN model from commonly used frameworks to a 16-bit microcontroller, and (2) automatically implements the model on the microcontroller to perform on-device inference. The optimization of data transfer saves time and mini-mizes chances of error, and the automatic implementation reduces the complexity to implement DNNs and CNNs on ultra-low-power microcontrollers.
Deep learning on climatic data holds potential for macroecological applications. However, its adoption remains limited among scientists outside the deep learning community due to storage, compute, and technical expertise barriers. To address this, we introduce Climplicit, a spatio-temporal geolocation encoder pretrained to generate implicit climatic representations anywhere on Earth. By bypassing the need to download raw climatic rasters and train feature extractors, our model uses x1000 fewer disk space and significantly reduces computational needs for downstream tasks. We evaluate our Climplicit embeddings on biomes classification, species distribution modeling, and plant trait regression. We find that linear probing our Climplicit embeddings consistently performs better or on par with training a model from scratch on downstream tasks and overall better than alternative geolocation encoding models.
The prospect of 4D video in Extended Reality (XR) platform is huge and exciting, it opens a whole new way of human computer interaction and the way we perceive the reality and consume multimedia. In this thesis, we have shown that feasibility of rendering 4D video in Microsoft mixed reality platform. This enables us to port any 3D performance capture from CVSSP into XR product like the HoloLens device with relative ease. However, if the 3D model is too complex and is made up of millions of vertices, the data bandwidth required to port the model is a severe limitation with the current hardware and communication system. Therefore, in this project we have also developed a compact representation of both shape and appearance of the 4d video sequence using deep learning models to effectively learn the compact representation of 4D video sequence and reconstruct it without affecting the shape and appearance of the video sequence.
Darknet markets (DNMs) facilitate the trade of illegal goods on a global scale. Gathering data on DNMs is critical to ensuring law enforcement agencies can effectively combat crime. Manually extracting data from DNMs is an error-prone and time-consuming task. Aiming to automate this process we develop a framework for extracting data from DNMs and evaluate the application of three state-of-the-art Named Entity Recognition (NER) models, ELMo-BiLSTM \citep{ShahEtAl2022}, UniversalNER \citep{ZhouEtAl2024}, and GLiNER \citep{ZaratianaEtAl2023}, at the task of extracting complex entities from DNM product listing pages. We propose a new annotated dataset, which we use to train, fine-tune, and evaluate the models. Our findings show that state-of-the-art NER models perform well in information extraction from DNMs, achieving 91% Precision, 96% Recall, and an F1 score of 94%. In addition, fine-tuning enhances model performance, with UniversalNER achieving the best performance.
Forest stands are the fundamental units in forest management inventories, silviculture, and financial analysis within operational forestry. Over the past two decades, a common method for mapping stand borders has involved delineation through manual interpretation of stereographic aerial images. This is a time-consuming and subjective process, limiting operational efficiency and introducing inconsistencies. Substantial effort has been devoted to automating the process, using various algorithms together with aerial images and canopy height models constructed from airborne laser scanning (ALS) data, but manual interpretation remains the preferred method. Deep learning (DL) methods have demonstrated great potential in computer vision, yet their application to forest stand delineation remains unexplored in published research. This study presents a novel approach, framing stand delineation as a multiclass segmentation problem and applying a U-Net based DL framework. The model was trained and evaluated using multispectral images, ALS data, and an existing stand map created by an expert interpreter. Performance was assessed on independent data using overall accuracy, a standard metric for classification tasks that measures the proportions of correctly classified pixels. The model achieved an overall accuracy of 0.73. These results demonstrate strong potential for DL in automated stand delineation. However, a few key challenges were noted, especially for complex forest environments.