Front-end electronics equipped with high-speed digitizers are being used and proposed for future nuclear detectors. Recent literature reveals that deep learning models, especially one-dimensional convolutional neural networks, are promising when dealing with digital signals from nuclear detectors. Simulations and experiments demonstrate the satisfactory accuracy and additional benefits of neural networks in this area. However, specific hardware accelerating such models for online operations still needs to be studied. In this work, we introduce PulseDL-II, a system-on-chip (SoC) specially designed for applications of event feature (time, energy, etc.) extraction from pulses with deep learning. Based on the previous version, PulseDL-II incorporates a RISC CPU into the system structure for better functional flexibility and integrity. The neural network accelerator in the SoC adopts a three-level (arithmetic unit, processing element, neural network) hierarchical architecture and facilitates parameter optimization of the digital design. Furthermore, we devise a quantization scheme compatible with deep learning frameworks (e.g., TensorFlow) within a selected subset of layer types. We validate the correct operations of PulseDL-II on field programmable gate arrays (FPGA) alone and with an experimental setup comprising a direct digital synthesis (DDS) and analog-to-digital converters (ADC). The proposed system achieved 60 ps time resolution and 0.40% energy resolution at signal to noise ratio (SNR) of 47.4 dB.
Index funds are substantially preferred by investors nowadays, and market sensitivities are instrumental in managing index funds. An index fund is a mutual fund aiming to track the returns of a predefined market index (e.g., the S&P 500). A basic strategy to manage an index fund is replicating the index's constituents and weights identically, which is, however, cost-ineffective and impractical. To address this issue, it is required to replicate the index partially with accurately predicted market sensitivities. Accordingly, we propose a novel partial-replication method via learning to predict market sensitivities. We first examine deep-learning models to predict market sensitivities in a supervised manner with our data-processing methods. Then, we propose a partial-index-tracking optimization model controlling the net predicted market sensitivities of the portfolios and index to be the same. These processes' efficacy is corroborated by our experiments on the Korea Stock Price Index 200. Our experiments show a significant reduction of the prediction errors compared with historical estimations and competitive tracking errors of replicating the index utilizing fewer than half of the entire constituents. Therefore, we show that applying deep learning to predict market sensitivities is promising and that our portfolio construction methods are practically effective. Additionally, to our knowledge, this is the first study addressing market sensitivities focused on deep learning.
Classifying Electroencephalogram(EEG) signals helps in understanding Brain-Computer Interface (BCI). EEG signals are vital in studying how the human mind functions. In this paper, we have used an Arithmetic Calculation dataset consisting of Before Calculation Signals (BCS) and During Calculation Signals (DCS). The dataset consisted of 36 participants. In order to understand the functioning of neurons in the brain, we classified BCS vs DCS. For this classification, we extracted various features such as Mutual Information (MI), Phase Locking Value (PLV), and Entropy namely Permutation entropy, Spectral entropy, Singular value decomposition entropy, Approximate entropy, Sample entropy. The classification of these features was done using RNN-based classifiers such as LSTM, BLSTM, ConvLSTM, and CNN-LSTM. The model achieved an accuracy of 99.72% when entropy was used as a feature and ConvLSTM as a classifier.
In coming years or decades, artificial general intelligence (AGI) may surpass human capabilities across many critical domains. We argue that, without substantial effort to prevent it, AGIs could learn to pursue goals that are in conflict (i.e. misaligned) with human interests. If trained like today's most capable models, AGIs could learn to act deceptively to receive higher reward, learn misaligned internally-represented goals which generalize beyond their fine-tuning distributions, and pursue those goals using power-seeking strategies. We review emerging evidence for these properties. In this revised paper, we include more direct empirical evidence published as of early 2025. AGIs with these properties would be difficult to align and may appear aligned even when they are not. Finally, we briefly outline how the deployment of misaligned AGIs might irreversibly undermine human control over the world, and we review research directions aimed at preventing this outcome.
Deep neural networks use multiple layers of functions to map an object represented by an input vector progressively to different representations, and with sufficient training, eventually to a single score for each class that is the output of the final decision function. Ideally, in this output space, the objects of different classes achieve maximum separation. Motivated by the need to better understand the inner working of a deep neural network, we analyze the effectiveness of the learned representations in separating the classes from a data complexity perspective. Using a simple complexity measure, a popular benchmarking task, and a well-known architecture design, we show how the data complexity evolves through the network, how it changes during training, and how it is impacted by the network design and the availability of training samples. We discuss the implications of the observations and the potentials for further studies.
Clinical Natural Language Processing (NLP) has become an emerging technology in healthcare that leverages a large amount of free-text data in electronic health records (EHRs) to improve patient care, support clinical decisions, and facilitate clinical and translational science research. Recently, deep learning has achieved state-of-the-art performance in many clinical NLP tasks. However, training deep learning models usually requires large annotated datasets, which are normally not publicly available and can be time-consuming to build in clinical domains. Working with smaller annotated datasets is typical in clinical NLP and therefore, ensuring that deep learning models perform well is crucial for the models to be used in real-world applications. A widely adopted approach is fine-tuning existing Pre-trained Language Models (PLMs), but these attempts fall short when the training dataset contains only a few annotated samples. Few-Shot Learning (FSL) has recently been investigated to tackle this problem. Siamese Neural Network (SNN) has been widely utilized as an FSL approach in computer vision, but has not been studied well in NLP. Furthermore, the literature on its applications in clinical domains is scarce. In this paper, we propose two SNN-based FSL approaches for clinical NLP, including Pre-Trained SNN (PT-SNN) and SNN with Second-Order Embeddings (SOE-SNN). We evaluated the proposed approaches on two clinical tasks, namely clinical text classification and clinical named entity recognition. We tested three few-shot settings including 4-shot, 8-shot, and 16-shot learning. Both clinical NLP tasks were benchmarked using three PLMs, including BERT,BioBERT, and BioClinicalBERT. The experimental results verified the effectiveness of the proposed SNN-based FSL approaches in both NLP tasks.
The use of Environmental Microorganisms (EMs) offers a highly efficient, low cost and harmless remedy to environmental pollution, by monitoring and decomposing of pollutants. This relies on how the EMs are correctly segmented and identified. With the aim of enhancing the segmentation of weakly visible EM images which are transparent, noisy and have low contrast, a Pairwise Deep Learning Feature Network (PDLF-Net) is proposed in this study. The use of PDLFs enables the network to focus more on the foreground (EMs) by concatenating the pairwise deep learning features of each image to different blocks of the base model SegNet. Leveraging the Shi and Tomas descriptors, we extract each image's deep features on the patches, which are centered at each descriptor using the VGG-16 model. Then, to learn the intermediate characteristics between the descriptors, pairing of the features is performed based on the Delaunay triangulation theorem to form pairwise deep learning features. In this experiment, the PDLF-Net achieves outstanding segmentation results of 89.24%, 63.20%, 77.27%, 35.15%, 89.72%, 91.44% and 89.30% on the accuracy, IoU, Dice, VOE, sensitivity, precision and specificity, respectively.
We aim to resolve this problem by introducing a comprehensive distributed deep learning (DDL) profiler, which can determine the various execution "stalls" that DDL suffers from while running on a public cloud. We have implemented the profiler by extending prior work to additionally estimate two types of communication stalls - interconnect and network stalls. We train popular DNN models using the profiler to characterize various AWS GPU instances and list their advantages and shortcomings for users to make an informed decision. We observe that the more expensive GPU instances may not be the most performant for all DNN models and AWS may sub-optimally allocate hardware interconnect resources. Specifically, the intra-machine interconnect can introduce communication overheads up to 90% of DNN training time and network-connected instances can suffer from up to 5x slowdown compared to training on a single instance. Further, we model the impact of DNN macroscopic features such as the number of layers and the number of gradients on communication stalls. Finally, we propose a measurement-based recommendation model for users to lower their public cloud monetary costs for DDL, given a time budget.
The traditional homography estimation pipeline consists of four main steps: feature detection, feature matching, outlier removal and transformation estimation. Recent deep learning models intend to address the homography estimation problem using a single convolutional network. While these models are trained in an end-to-end fashion to simplify the homography estimation problem, they lack the feature matching step and/or the outlier removal step, which are important steps in the traditional homography estimation pipeline. In this paper, we attempt to build a deep learning model that mimics all four steps in the traditional homography estimation pipeline. In particular, the feature matching step is implemented using the cost volume technique. To remove outliers in the cost volume, we treat this outlier removal problem as a denoising problem and propose a novel self-supervised loss to solve the problem. Extensive experiments on synthetic and real datasets demonstrate that the proposed model outperforms existing deep learning models.
With the rapid advancement and increased use of deep learning models in image identification, security becomes a major concern to their deployment in safety-critical systems. Since the accuracy and robustness of deep learning models are primarily attributed from the purity of the training samples, therefore the deep learning architectures are often susceptible to adversarial attacks. Adversarial attacks are often obtained by making subtle perturbations to normal images, which are mostly imperceptible to humans, but can seriously confuse the state-of-the-art machine learning models. We propose a framework, named APuDAE, leveraging Denoising AutoEncoders (DAEs) to purify these samples by using them in an adaptive way and thus improve the classification accuracy of the target classifier networks that have been attacked. We also show how using DAEs adaptively instead of using them directly, improves classification accuracy further and is more robust to the possibility of designing adaptive attacks to fool them. We demonstrate our results over MNIST, CIFAR-10, ImageNet dataset and show how our framework (APuDAE) provides comparable and in most cases better performance to the baseline methods in purifying adversaries. We also design adaptive attack specifically designed to attack our purifying model and demonstrate how our defense is robust to that.
Diagnosing erythema migrans (EM) skin lesion, the most common early symptom of Lyme disease using deep learning techniques can be effective to prevent long-term complications. Existing works on deep learning based EM recognition only utilizes lesion image due to the lack of a dataset of Lyme disease related images with associated patient data. Physicians rely on patient information about the background of the skin lesion to confirm their diagnosis. In order to assist the deep learning model with a probability score calculated from patient data, this study elicited opinion from fifteen doctors. For the elicitation process, a questionnaire with questions and possible answers related to EM was prepared. Doctors provided relative weights to different answers to the questions. We converted doctors evaluations to probability scores using Gaussian mixture based density estimation. For elicited probability model validation, we exploited formal concept analysis and decision tree. The elicited probability scores can be utilized to make image based deep learning Lyme disease pre-scanners robust.
First Rosenblatt's theorem about omnipotence of shallow networks states that elementary perceptrons can solve any classification problem if there are no discrepancies in the training set. Minsky and Papert considered elementary perceptrons with restrictions on the neural inputs: a bounded number of connections or a relatively small diameter of the receptive field for each neuron at the hidden layer. They proved that under these constraints, an elementary perceptron cannot solve some problems, such as the connectivity of input images or the parity of pixels in them. In this note, we demonstrated first Rosenblatt's theorem at work, showed how an elementary perceptron can solve a version of the travel maze problem, and analysed the complexity of that solution. We constructed also a deep network algorithm for the same problem. It is much more efficient. The shallow network uses an exponentially large number of neurons on the hidden layer (Rosenblatt's $A$-elements), whereas for the deep network the second order polynomial complexity is sufficient. We demonstrated that for the same complex problem deep network can be much smaller and reveal a heuristic behind this effect.
Data redundancy is ubiquitous in the inputs and intermediate results of Deep Neural Networks (DNN). It offers many significant opportunities for improving DNN performance and efficiency and has been explored in a large body of work. These studies have scattered in many venues across several years. The targets they focus on range from images to videos and texts, and the techniques they use to detect and exploit data redundancy also vary in many aspects. There is not yet a systematic examination and summary of the many efforts, making it difficult for researchers to get a comprehensive view of the prior work, the state of the art, differences and shared principles, and the areas and directions yet to explore. This article tries to fill the void. It surveys hundreds of recent papers on the topic, introduces a novel taxonomy to put the various techniques into a single categorization framework, offers a comprehensive description of the main methods used for exploiting data redundancy in improving multiple kinds of DNNs on data, and points out a set of research opportunities for future to explore.
Deep learning based computer vision models are increasingly used by urban planners to support decision making for shaping urban environments. Such models predict how people perceive the urban environment quality in terms of e.g. its safety or beauty. However, the blackbox nature of deep learning models hampers urban planners to understand what landscape objects contribute to a particularly high quality or low quality urban space perception. This study investigates how computer vision models can be used to extract relevant policy information about peoples' perception of the urban space. To do so, we train two widely used computer vision architectures; a Convolutional Neural Network and a transformer, and apply GradCAM -- a well-known ex-post explainable AI technique -- to highlight the image regions important for the model's prediction. Using these GradCAM visualizations, we manually annotate the objects relevant to the models' perception predictions. As a result, we are able to discover new objects that are not represented in present object detection models used for annotation in previous studies. Moreover, our methodological results suggest that transformer architectures are better suited to be used in combination with GradCAM techniques. Code is available on Github.
Ammunition scrap inspection is an essential step in the process of recycling ammunition metal scrap. Most ammunition is composed of a number of components, including case, primer, powder, and projectile. Ammo scrap containing energetics is considered to be potentially dangerous and should be separated before the recycling process. Manually inspecting each piece of scrap is tedious and time-consuming. We have gathered a dataset of ammunition components with the goal of applying artificial intelligence for classifying safe and unsafe scrap pieces automatically. First, two training datasets are manually created from visual and x-ray images of ammo. Second, the x-ray dataset is augmented using the spatial transforms of histogram equalization, averaging, sharpening, power law, and Gaussian blurring in order to compensate for the lack of sufficient training data. Lastly, the representative YOLOv4 object detection method is applied to detect the ammo components and classify the scrap pieces into safe and unsafe classes, respectively. The trained models are tested against unseen data in order to evaluate the performance of the applied method. The experiments demonstrate the feasibility of ammo component detection and classification using deep learning. The datasets and the pre-trained models are available at https://github.com/hadi-ghnd/Scrap-Classification.
Accurate prediction of crop yield before harvest is of great importance for crop logistics, market planning, and food distribution around the world. Yield prediction requires monitoring of phenological and climatic characteristics over extended time periods to model the complex relations involved in crop development. Remote sensing satellite images provided by various satellites circumnavigating the world are a cheap and reliable way to obtain data for yield prediction. The field of yield prediction is currently dominated by Deep Learning approaches. While the accuracies reached with those approaches are promising, the needed amounts of data and the ``black-box'' nature can restrict the application of Deep Learning methods. The limitations can be overcome by proposing a pipeline to process remote sensing images into feature-based representations that allow the employment of Extreme Gradient Boosting (XGBoost) for yield prediction. A comparative evaluation of soybean yield prediction within the United States shows promising prediction accuracies compared to state-of-the-art yield prediction systems based on Deep Learning. Feature importances expose the near-infrared spectrum of light as an important feature within our models. The reported results hint at the capabilities of XGBoost for yield prediction and encourage future experiments with XGBoost for yield prediction on other crops in regions all around the world.
Gradient-based deep-learning algorithms exhibit remarkable performance in practice, but it is not well-understood why they are able to generalize despite having more parameters than training examples. It is believed that implicit bias is a key factor in their ability to generalize, and hence it was widely studied in recent years. In this short survey, we explain the notion of implicit bias, review main results and discuss their implications.
The standard diagnostic procedures for targeted therapies in lung cancer treatment involve histological subtyping and subsequent detection of key driver mutations, such as EGFR. Even though molecular profiling can uncover the driver mutation, the process is often expensive and time-consuming. Deep learning-oriented image analysis offers a more economical alternative for discovering driver mutations directly from whole slide images (WSIs). In this work, we used customized deep learning pipelines with weak supervision to identify the morphological correlates of EGFR mutation from hematoxylin and eosin-stained WSIs, in addition to detecting tumor and histologically subtyping it. We demonstrate the effectiveness of our pipeline by conducting rigorous experiments and ablation studies on two lung cancer datasets - TCGA and a private dataset from India. With our pipeline, we achieved an average area under the curve (AUC) of 0.964 for tumor detection, and 0.942 for histological subtyping between adenocarcinoma and squamous cell carcinoma on the TCGA dataset. For EGFR detection, we achieved an average AUC of 0.864 on the TCGA dataset and 0.783 on the dataset from India. Our key learning points include the following. Firstly, there is no particular advantage of using a feature extractor layers trained on histology, if one is going to fine-tune the feature extractor on the target dataset. Secondly, selecting patches with high cellularity, presumably capturing tumor regions, is not always helpful, as the sign of a disease class may be present in the tumor-adjacent stroma.
The joint optimization of the reconstruction and classification error is a hard non convex problem, especially when a non linear mapping is utilized. In order to overcome this obstacle, a novel optimization strategy is proposed, in which a Convolutional Autoencoder for dimensionality reduction and a classifier composed by a Fully Connected Network, are combined to simultaneously produce supervised dimensionality reduction and predictions. It turned out that this methodology can also be greatly beneficial in enforcing explainability of deep learning architectures. Additionally, the resulting Latent Space, optimized for the classification task, can be utilized to improve traditional, interpretable classification algorithms. The experimental results, showed that the proposed methodology achieved competitive results against the state of the art deep learning methods, while being much more efficient in terms of parameter count. Finally, it was empirically justified that the proposed methodology introduces advanced explainability regarding, not only the data structure through the produced latent space, but also about the classification behaviour.
In some areas, such as Cognitive Linguistics, researchers are still using traditional techniques based on manual rules and patterns. Since the definition of shell noun is rather subjective and there are many exceptions, this time-consuming work had to be done by hand in the past when Deep Learning techniques were not mature enough. With the increasing number of networked languages, these rules are becoming less useful. However, there is a better alternative now. With the development of Deep Learning, pre-trained language models have provided a good technical basis for Natural Language Processing. Automated processes based on Deep Learning approaches are more in line with modern needs. This paper collaborates across borders to propose two Neural Network models for the automatic detection of shell nouns and experiment on the WikiText-2 dataset. The proposed approaches not only allow the entire process to be automated, but the precision has reached 94% even on completely unseen articles, comparable to that of human annotators. This shows that the performance and generalization ability of the model is good enough to be used for research purposes. Many new nouns are found that fit the definition of shell noun very well. All discovered shell nouns as well as pre-trained models and code are available on GitHub.