Objective: Bloch simulation constitutes an essential part of magnetic resonance imaging (MRI) development. However, even with the graphics processing unit (GPU) acceleration, the heavy computational load remains a major challenge, especially in large-scale, high-accuracy simulation scenarios. This work aims to develop a deep learning-based simulator to accelerate Bloch simulation. Approach: The simulator model, called Simu-Net, is based on an end-to-end convolutional neural network and is trained with synthetic data generated by traditional Bloch simulation. It uses dynamic convolution to fuse spatial and physical information with different dimensions and introduces position encoding templates to achieve position-specific labeling and overcome the receptive field limitation of the convolutional network. Main Results: Compared with mainstream GPU-based MRI simulation software, Simu-Net successfully accelerates simulations by hundreds of times in both traditional and advanced MRI pulse sequences. The accuracy and robustness of the proposed framework were verified qualitatively and quantitatively. Besides, the trained Simu-Net was applied to generate sufficient customized training samples for deep learning-based T2 mapping and comparable results to conventional methods were obtained in the human brain. Significance: As a proof-of-concept work, Simu-Net shows the potential to apply deep learning for rapidly approximating the forward physical process of MRI and may increase the efficiency of Bloch simulation for optimization of MRI pulse sequences and deep learning-based methods.
We study the application of large language models to zero-shot and few-shot classification of tabular data. We prompt the large language model with a serialization of the tabular data to a natural-language string, together with a short description of the classification problem. In the few-shot setting, we fine-tune the large language model using some labeled examples. We evaluate several serialization methods including templates, table-to-text models, and large language models. Despite its simplicity, we find that this technique outperforms prior deep-learning-based tabular classification methods on several benchmark datasets. In most cases, even zero-shot classification obtains non-trivial performance, illustrating the method's ability to exploit prior knowledge encoded in large language models. Unlike many deep learning methods for tabular datasets, this approach is also competitive with strong traditional baselines like gradient-boosted trees, especially in the very-few-shot setting.
Deep learning (DL) based channel state information (CSI) feedback in multiple-input multiple-output (MIMO) systems recently has attracted lots of attention from both academia and industrial. From a practical point of views, it is huge burden to train, transfer and deploy a DL model for each parameter configuration of the base station (BS). In this paper, we propose a scalable and flexible framework for DL based CSI feedback referred as scalable CsiNet (SCsiNet) to adapt a family of configured parameters such as feedback payloads, MIMO channel ranks, antenna numbers. To reduce model size and training complexity, the core block with pre-processing and post-processing in SCsiNet is reused among different parameter configurations as much as possible which is totally different from configuration-orienting design. The preprocessing and post-processing are trainable neural network layers introduced for matching input/output dimensions and probability distributions. The proposed SCsiNet is evaluated by metrics of squared generalized cosine similarity (SGCS) and user throughput (UPT) in system level simulations. Compared to existing schemes (configuration-orienting DL schemes and 3GPP Rel-16 Type-II codebook based schemes), the proposed scheme can significantly reduce mode size and achieve 2%-10% UPT improvement for all parameter configurations.
When a deep learning model is sequentially trained on different datasets, it forgets the knowledge acquired from previous data, a phenomenon known as catastrophic forgetting. It deteriorates performance of the deep learning model on diverse datasets, which is critical in privacy-preserving deep learning (PPDL) applications based on transfer learning (TL). To overcome this, we propose review learning (RL), a generative-replay-based continual learning technique that does not require a separate generator. Data samples are generated from the memory stored within the synaptic weights of the deep learning model which are used to review knowledge acquired from previous datasets. The performance of RL was validated through PPDL experiments. Simulations and real-world medical multi-institutional experiments were conducted using three types of binary classification electronic health record data. In the real-world experiments, the global area under the receiver operating curve was 0.710 for RL and 0.655 for TL. Thus, RL was highly effective in retaining previously learned knowledge.
As deep learning models nowadays are widely adopted by both cloud services and edge devices, reducing the latency of deep learning model inferences becomes crucial to provide efficient model serving. However, it is challenging to develop efficient tensor programs for deep learning operators due to the high complexity of modern accelerators and the rapidly growing number of operators. Deep learning compilers, such as Apache TVM, adopt declarative scheduling primitives to lower the bar of developing tensor programs. However, we show that this approach is insufficient to cover state-of-the-art tensor program optimizations. In this paper, we propose to embed the scheduling process into tensor programs and use dedicated mappings, called task mappings, to define the computation assignment and ordering. This new approach greatly enriches the expressible optimizations by allowing developers to manipulate tensor programs at a much finer granularity. We call the proposed method the task-mapping programming paradigm. In addition, we propose a new post-scheduling fusion optimization that allows developers to focus on scheduling every single operator and automates the fusion after scheduling. It greatly reduces the engineering efforts for operator fusion. Our proposed paradigm also constructs an efficient hardware-centric schedule space, which is agnostic to the program input size and greatly reduces the tuning time. With the proposed paradigm, we implement a deep learning compiler Hidet. Extensive experiments on modern convolution and transformer models show that Hidet outperforms state-of-the-art DNN inference framework, ONNX Runtime, and compiler, TVM equipped with scheduler AutoTVM and Ansor, by up to 1.48x (1.22x on average). It also reduces the tuning time by 20x and 11x compared with AutoTVM and Ansor, respectively. We open-sourced hidet at https://www.github.com/hidet-org/hidet.
Parameter estimation and associated uncertainty quantification is an important problem in dynamical systems characterized by ordinary differential equation (ODE) models that are often nonlinear. Typically, such models have analytically intractable trajectories which result in likelihoods and posterior distributions that are similarly intractable. Bayesian inference for ODE systems via simulation methods require numerical approximations to produce inference with high accuracy at a cost of heavy computational power and slow convergence. At the same time, Artificial Neural Networks (ANN) offer tractability that can be utilized to construct an approximate but tractable likelihood and posterior distribution. In this paper we propose a hybrid approach, where Laplace-based Bayesian inference is combined with an ANN architecture for obtaining approximations to the ODE trajectories as a function of the unknown initial values and system parameters. Suitable choices of a collocation grid and customized loss functions are proposed to fine tune the ODE trajectories and Laplace approximation. The effectiveness of our proposed methods is demonstrated using an epidemiological system with non-analytical solutions, the Susceptible-Infectious-Removed (SIR) model for infectious diseases, based on simulated and real-life influenza datasets. The novelty and attractiveness of our proposed approach include (i) a new development of Bayesian inference using ANN architectures for ODE based dynamical systems, and (ii) a computationally fast posterior inference by avoiding convergence issues of benchmark Markov Chain Monte Carlo methods. These two features establish the developed approach as an accurate alternative to traditional Bayesian computational methods, with improved computational cost.
The ability to generate synthetic data has a variety of use cases across different domains. In education research, there is a growing need to have access to synthetic data to test certain concepts and ideas. In recent years, several deep learning architectures were used to aid in the generation of synthetic data but with varying results. In the education context, the sophistication of implementing different models requiring large datasets is becoming very important. This study aims to compare the application of synthetic tabular data generation between a probabilistic model specifically a Bayesian Network, and a deep learning model, specifically a Generative Adversarial Network using a classification task. The results of this study indicate that synthetic tabular data generation is better suited for the education context using probabilistic models (overall accuracy of 75%) than deep learning architecture (overall accuracy of 38%) because of probabilistic interdependence. Lastly, we recommend that other data types, should be explored and evaluated for their application in generating synthetic data for education use cases.
Background: Object detection in radiograph computer vision has largely benefited from progress in deep convolutional neural networks and can, for example, annotate a radiograph with a box around a knee joint or intervertebral disc. Is deep learning capable of detect small (less than 1% of the image) in radiographs? And how many radiographs do we need use when re-training a deep learning model? Methods: We annotated 396 radiographs of left and right carpi dorsal 75 medial to palmarolateral oblique (DMPLO) projection with the location of radius, proximal row of carpal bones, distal row of carpal bones, accessory carpal bone, first carpal bone (if present), and metacarpus (metacarpal II, III, and IV). The radiographs and respective annotations were splited into sets that were used to leave-one-out cross-validation of models created using transfer learn from YOLOv5s. Results: Models trained using 96 radiographs or more achieved precision, recall and mAP above 0.95, including for the first carpal bone, when trained for 32 epochs. The best model needed the double of epochs to learn to detect the first carpal bone compared with the other bones. Conclusions: Free and open source state of the art object detection models based on deep learning can be re-trained for radiograph computer vision applications with 100 radiographs and achieved precision, recall and mAP above 0.95.
Currently, a significant amount of research has been done in field of Remote Sensing with the use of deep learning techniques. The introduction of Marine Debris Archive (MARIDA), an open-source dataset with benchmark results, for marine debris detection opened new pathways to use deep learning techniques for the task of debris detection and segmentation. This paper introduces a novel attention based segmentation technique that outperforms the existing state-of-the-art results introduced with MARIDA. The paper presents a novel spatial aware encoder and decoder architecture to maintain the contextual information and structure of sparse ground truth patches present in the images. The attained results are expected to pave the path for further research involving deep learning using remote sensing images. The code is available at https://github.com/sheikhazhanmohammed/SADMA.git
The interpretability of neural networks has recently received extensive attention. Previous prototype-based explainable networks involved prototype activation in both reasoning and interpretation processes, requiring specific explainable structures for the prototype, thus making the network less accurate as it gains interpretability. Therefore, the decoupling prototypical network (DProtoNet) was proposed to avoid this problem. This new model contains encoder, inference, and interpretation modules. As regards the encoder module, unrestricted feature masks were presented to generate expressive features and prototypes. Regarding the inference module, a multi-image prototype learning method was introduced to update prototypes so that the network can learn generalized prototypes. Finally, concerning the interpretation module, a multiple dynamic masks (MDM) decoder was suggested to explain the neural network, which generates heatmaps using the consistent activation of the original image and mask image at the detection nodes of the network. It decouples the inference and interpretation modules of a prototype-based network by avoiding the use of prototype activation to explain the network's decisions in order to simultaneously improve the accuracy and interpretability of the neural network. The multiple public general and medical datasets were tested, and the results confirmed that our method could achieve a 5% improvement in accuracy and state-of-the-art interpretability compared with previous methods.
During the investigation of criminal activity when evidence is available, the issue at hand is determining the credibility of the video and ascertaining that the video is real. Today, one way to authenticate the footage is to identify the camera that was used to capture the image or video in question. While a very common way to do this is by using image meta-data, this data can easily be falsified by changing the video content or even splicing together content from two different cameras. Given the multitude of solutions proposed to this problem, it is yet to be sufficiently solved. The aim of our project is to build an algorithm that identifies which camera was used to capture an image using traces of information left intrinsically in the image, using filters, followed by a deep neural network on these filters. Solving this problem would have a big impact on the verification of evidence used in criminal and civil trials and even news reporting.
Dysgraphia, a handwriting learning disability, has a serious negative impact on children's academic results, daily life and overall wellbeing. Early detection of dysgraphia allows for an early start of a targeted intervention. Several studies have investigated dysgraphia detection by machine learning algorithms using a digital tablet. However, these studies deployed classical machine learning algorithms with manual feature extraction and selection as well as binary classification: either dysgraphia or no dysgraphia. In this work, we investigated fine grading of handwriting capabilities by predicting SEMS score (between 0 and 12) with deep learning. Our approach provide accuracy more than 99% and root mean square error lower than one, with automatic instead of manual feature extraction and selection. Furthermore, we used smart pen called SensoGrip, a pen equipped with sensors to capture handwriting dynamics, instead of a tablet, enabling writing evaluation in more realistic scenarios.
Applying graph-based approaches in deep learning receives more attention over time. This study presents statistical analysis on the use of graph-based approaches in deep learning and examines the scientific impact of the related articles. Processing the data obtained from the Web of Science database, metrics such as the type of the articles, funding availability, indexing type, annual average number of citations and the number of access were analyzed to quantitatively reveal the effects on the scientific audience. It's outlined that deep learning-based studies gained momentum after year 2013, and the rate of graph-based approaches in all deep learning studies increased linearly from 1% to 4% within the following 10 years. Conference publications scanned in the Conference Proceeding Citation Index (CPCI) on the graph-based approaches receive significantly more citations. The citation counts of the SCI-Expanded and Emerging SCI indexed publications of the two streams are close to each other. While the citation performances of the supported and unsupported publications of the two sides were similar, pure deep learning studies received more citations on the journal publication side and graph-based approaches received more citations on the conference side. Despite their similar performance in recent years, graph-based studies show twice more citation performance as they get older, compared to traditional approaches. Annual average citation performance per article for all deep learning studies is 11.051 in 2014, while it is 22.483 for graph-based studies. Also, despite receiving 16% more access, graph-based papers get almost the same overall citation over time with the pure counterpart. This is an indication that graph-based approaches need a greater bunch of attention to follow, while pure deep learning counterpart is relatively simpler to get inside.
The automatic classification of radar waveform is a fundamental technique in electronic countermeasures (ECM).Recent supervised deep learning-based methods have achieved great success in a such classification task.However, those methods require enough labeled samples to work properly and in many circumstances, it is not available.To tackle this problem, in this paper, we propose a three-stages deep radar waveform clustering(DRSC) technique to automatically group the received signal samples without labels.Firstly, a pretext model is trained in a self-supervised way with the help of several data augmentation techniques to extract the class-dependent features.Next,the pseudo-supervised contrastive training is involved to further promote the separation between the extracted class-dependent features.And finally, the unsupervised problem is converted to a semi-supervised classification problem via pseudo label generation. The simulation results show that the proposed algorithm can effectively extract class-dependent features, outperforming several unsupervised clustering methods, even reaching performance on par with the supervised deep learning-based methods.
Although deep learning has made great progress in recent years, the exploding economic and environmental costs of training neural networks are becoming unsustainable. To address this problem, there has been a great deal of research on *algorithmically-efficient deep learning*, which seeks to reduce training costs not at the hardware or implementation level, but through changes in the semantics of the training program. In this paper, we present a structured and comprehensive overview of the research in this field. First, we formalize the *algorithmic speedup* problem, then we use fundamental building blocks of algorithmically efficient training to develop a taxonomy. Our taxonomy highlights commonalities of seemingly disparate methods and reveals current research gaps. Next, we present evaluation best practices to enable comprehensive, fair, and reliable comparisons of speedup techniques. To further aid research and applications, we discuss common bottlenecks in the training pipeline (illustrated via experiments) and offer taxonomic mitigation strategies for them. Finally, we highlight some unsolved research challenges and present promising future directions.
Deep learning models are gaining popularity and potency in predicting polymer properties. These models can be built using pre-existing data and are useful for the rapid prediction of polymer properties. However, the performance of a deep learning model is intricately connected to its topology and the volume of training data. There is no facile protocol available to select a deep learning architecture, and there is a lack of a large volume of homogeneous sequence-property data of polymers. These two factors are the primary bottleneck for the efficient development of deep learning models. Here we assess the severity of these factors and propose new algorithms to address them. We show that a linear layer-by-layer expansion of a neural network can help in identifying the best neural network topology for a given problem. Moreover, we map the discrete sequence space of a polymer to a continuous one-dimensional latent space using a machine learning pipeline to identify minimal data points for building a universal deep learning model. We implement these approaches for three representative cases of building sequence-property surrogate models, viz., the single-molecule radius of gyration of a copolymer, adhesive free energy of a copolymer, and copolymer compatibilizer, demonstrating the generality of the proposed strategies. This work establishes efficient methods for building universal deep learning models with minimal data and hyperparameters for predicting sequence-defined properties of polymers.
The normalizing layer has become one of the basic configurations of deep learning models, but it still suffers from computational inefficiency, interpretability difficulties, and low generality. After gaining a deeper understanding of the recent normalization and normalizer-free research works from a sample's perspective, we reveal the fact that the problem lies in the sampling noise and the inappropriate prior assumption. In this paper, we propose a simple and effective alternative to normalization, which is called "NoMorelization". NoMorelization is composed of two trainable scalars and a zero-centered noise injector. Experimental results demonstrate that NoMorelization is a general component for deep learning and is suitable for different model paradigms (e.g., convolution-based and attention-based models) to tackle different tasks (e.g., discriminative and generative tasks). Compared with existing mainstream normalizers (e.g., BN, LN, and IN) and state-of-the-art normalizer-free methods, NoMorelization shows the best speed-accuracy trade-off.
We present SELOR, a framework for integrating self-explaining capabilities into a given deep model to achieve both high prediction performance and human precision. By "human precision", we refer to the degree to which humans agree with the reasons models provide for their predictions. Human precision affects user trust and allows users to collaborate closely with the model. We demonstrate that logic rule explanations naturally satisfy human precision with the expressive power required for good predictive performance. We then illustrate how to enable a deep model to predict and explain with logic rules. Our method does not require predefined logic rule sets or human annotations and can be learned efficiently and easily with widely-used deep learning modules in a differentiable way. Extensive experiments show that our method gives explanations closer to human decision logic than other methods while maintaining the performance of deep learning models.
Despite tremendous progress in natural language processing using deep learning techniques in recent years, sign language production and comprehension has advanced very little. One critical barrier is the lack of largescale datasets available to the public due to the unbearable cost of labeled data generation. Efforts to provide public data for American Sign Language (ASL) comprehension have yielded two datasets, comprising more than thousand video clips. These datasets are large enough to enable a meaningful start to deep learning research on sign languages but are far too small to lead to any solution that can be practically deployed. So far, there is still no suitable dataset for ASL production. We proposed a system that can generate large scale ASL datasets for continuous ASL. It is suitable for general ASL processing and is particularly useful for ASL production. The continuous ASL dataset contains English labeled human articulations in condensed body pose data formats. To better serve the research community, we are releasing the first version of our ASL dataset, which contains 30k sentences, 416k words, a vocabulary of 18k words, in a total of 104 hours. This is the largest continuous sign language dataset published to date in terms of video duration. We also describe a system that can evolve and expand the dataset to incorporate better data processing techniques and more contents when available. It is our hope that the release of this ASL dataset and the sustainable dataset generation system to the public will propel better deep-learning research in ASL natural language processing.
Single-cell RNA-sequencing (scRNA-seq) has become a routinely used technique to quantify the gene expression profile of thousands of single cells simultaneously. Analysis of scRNA-seq data plays an important role in the study of cell states and phenotypes, and has helped elucidate biological processes, such as those occurring during development of complex organisms and improved our understanding of disease states, such as cancer, diabetes, and COVID, among others. Deep learning, a recent advance of artificial intelligence that has been used to address many problems involving large datasets, has also emerged as a promising tool for scRNA-seq data analysis, as it has a capacity to extract informative, compact features from noisy, heterogeneous, and high-dimensional scRNA-seq data to improve downstream analysis. The present review aims at surveying recently developed deep learning techniques in scRNA-seq data analysis, identifying key steps within the scRNA-seq data analysis pipeline that have been advanced by deep learning, and explaining the benefits of deep learning over more conventional analysis tools. Finally, we summarize the challenges in current deep learning approaches faced within scRNA-seq data and discuss potential directions for improvements in deep algorithms for scRNA-seq data analysis.