Convolutional Neural Networks (CNNs) constitute a class of Deep Learning models which have been used in the recent past to resolve many problems in computer vision, in particular optical flow estimation. Measuring displacement and strain fields can be regarded as a particular case of this problem. However, it seems that CNNs have never been used so far to perform such measurements. This work is aimed at implementing a CNN able to retrieve displacement and strain fields from pairs of reference and deformed images of a flat speckled surface, as Digital Image Correlation (DIC) does. This paper explains how a CNN called StrainNet can be developed to reach this goal, and how specific ground truth datasets are elaborated to train this CNN. The main result is that StrainNet successfully performs such measurements, and that it achieves competing results in terms of metrological performance and computing time. The conclusion is that CNNs like StrainNet offer a viable alternative to DIC, especially for real-time applications.
A generative model is a statistical model that is able to generate new data instances from previously observed ones. In the context of business processes, a generative model creates new execution traces from a set of historical traces, also known as an event log. Two families of generative process simulation models have been developed in previous work: data-driven simulation models and deep learning models. Until now, these two approaches have evolved independently and their relative performance has not been studied. This paper fills this gap by empirically comparing a data-driven simulation technique with multiple deep learning techniques, which construct models are capable of generating execution traces with timestamped events. The study sheds light into the relative strengths of both approaches and raises the prospect of developing hybrid approaches that combine these strengths.
Brief fragments of sleep shorter than 15 s are defined as microsleep episodes (MSEs), often subjectively perceived as sleepiness. Their main characteristic is a slowing in frequency in the electroencephalogram (EEG), similar to stage N1 sleep according to standard criteria. The maintenance of wakefulness test (MWT) is often used in a clinical setting to assess vigilance. Scoring of the MWT in most sleep-wake centers is limited to classical definition of sleep (30-s epochs), and MSEs are mostly not considered in the absence of established scoring criteria defining MSEs but also because of the laborious work. We aimed for automatic detection of MSEs with machine learning, i.e. with deep learning based on raw EEG and EOG data as input. We analyzed MWT data of 76 patients. Experts visually scored wakefulness, and according to recently developed scoring criteria MSEs, microsleep episode candidates (MSEc), and episodes of drowsiness (ED). We implemented segmentation algorithms based on convolutional neural networks (CNNs) and a combination of a CNN with a long-short term memory (LSTM) network. A LSTM network is a type of a recurrent neural network which has a memory for past events and takes them into account. Data of 53 patients were used for training of the classifiers, 12 for validation and 11 for testing. Our algorithms showed a good performance close to human experts. The detection was very good for wakefulness and MSEs and poor for MSEc and ED, similar to the low inter-expert reliability for these borderline segments. We provide a proof of principle that it is feasible to reliably detect MSEs with deep neuronal networks based on raw EEG and EOG data with a performance close to that of human experts. Code of algorithms ( https://github.com/alexander-malafeev/microsleep-detection ) and data ( https://zenodo.org/record/3251716 ) are available.
Arabic text recognition is a challenging task because of the cursive nature of Arabic writing system, its joint writing scheme, the large number of ligatures and many other challenges. Deep Learning DL models achieved significant progress in numerous domains including computer vision and sequence modelling. This paper presents a model that can recognize Arabic text that was printed using multiple font types including fonts that mimic Arabic handwritten scripts. The proposed model employs a hybrid DL network that can recognize Arabic printed text without the need for character segmentation. The model was tested on a custom dataset comprised of over two million word samples that were generated using 18 different Arabic font types. The objective of the testing process was to assess the model capability in recognizing a diverse set of Arabic fonts representing a varied cursive styles. The model achieved good results in recognizing characters and words and it also achieved promising results in recognizing characters when it was tested on unseen data. The prepared model, the custom datasets and the toolkit for generating similar datasets are made publicly available, these tools can be used to prepare models for recognizing other font types as well as to further extend and enhance the performance of the proposed model.
Structured Illumination Microscopy is a widespread methodology to image live and fixed biological structures smaller than the diffraction limits of conventional optical microscopy. Using recent advances in image up-scaling through deep learning models, we demonstrate a method to reconstruct 3D SIM image stacks with twice the axial resolution attainable through conventional SIM reconstructions. We further evaluate our method for robustness to noise & generalisability to varying observed specimens, and discuss potential adaptions of the method to further improvements in resolution.
Extracting behavioral measurements non-invasively from video is stymied by the fact that it is a hard computational problem. Recent advances in deep learning have tremendously advanced predicting posture from videos directly, which quickly impacted neuroscience and biology more broadly. In this primer we review the budding field of motion capture with deep learning. In particular, we will discuss the principles of those novel algorithms, highlight their potential as well as pitfalls for experimentalists, and provide a glimpse into the future.
Deep learning-based image reconstruction methods have achieved promising results across multiple MRI applications. However, most approaches require large-scale fully-sampled ground truth data for supervised training. Acquiring fully-sampled data is often either difficult or impossible, particularly for dynamic contrast enhancement (DCE), 3D cardiac cine, and 4D flow. We present a deep learning framework for MRI reconstruction without any fully-sampled data using generative adversarial networks. We test the proposed method in two scenarios: retrospectively undersampled fast spin echo knee exams and prospectively undersampled abdominal DCE. The method recovers more anatomical structure compared to conventional methods.
I propose a state of the art deep neural architectural solution for handwritten character recognition for Bengali alphabets, compound characters as well as numerical digits that achieves state-of-the-art accuracy 96.8% in just 11 epochs. Similar work has been done before by Chatterjee, Swagato, et al. but they achieved 96.12% accuracy in about 47 epochs. The deep neural architecture used in that paper was fairly large considering the inclusion of the weights of the ResNet 50 model which is a 50 layer Residual Network. This proposed model achieves higher accuracy as compared to any previous work & in a little number of epochs. ResNet50 is a good model trained on the ImageNet dataset, but I propose an HCR network that is trained from the scratch on Bengali characters without the "Ensemble Learning" that can outperform previous architectures.
Pollux improves scheduling performance in deep learning (DL) clusters by adaptively co-optimizing inter-dependent factors both at the per-job level and at the cluster-wide level. Most existing schedulers expect users to specify the number of resources for each job, often leading to inefficient resource use. Some recent schedulers choose job resources for users, but do so without awareness of how DL training can be re-optimized to better utilize the provided resources. Pollux simultaneously considers both aspects. By monitoring the status of each job during training, Pollux models how their goodput (a novel metric we introduce that combines system throughput with statistical efficiency) would change by adding or removing resources. Leveraging these information, Pollux dynamically (re-)assigns resources to improve cluster-wide goodput, while respecting fairness and continually optimizing each DL job to better utilize those resources. In experiments with real DL jobs and with trace-driven simulations, Pollux reduces average job completion times by 37-50% relative to state-of-the-art DL schedulers, even when they are provided with ideal resource and training configurations for every job. Pollux promotes fairness among DL jobs competing for resources based on a more meaningful measure of useful job progress, and reveals a new opportunity for reducing DL cost in cloud environments. Pollux is implemented and publicly available as part of an open-source project at https://github.com/petuum/adaptdl.
Nonlinear differential equations rarely admit closed-form solutions, thus requiring numerical time-stepping algorithms to approximate solutions. Further, many systems characterized by multiscale physics exhibit dynamics over a vast range of timescales, making numerical integration computationally expensive due to numerical stiffness. In this work, we develop a hierarchy of deep neural network time-steppers to approximate the flow map of the dynamical system over a disparate range of time-scales. The resulting model is purely data-driven and leverages features of the multiscale dynamics, enabling numerical integration and forecasting that is both accurate and highly efficient. Moreover, similar ideas can be used to couple neural network-based models with classical numerical time-steppers. Our multiscale hierarchical time-stepping scheme provides important advantages over current time-stepping algorithms, including (i) circumventing numerical stiffness due to disparate time-scales, (ii) improved accuracy in comparison with leading neural-network architectures, (iii) efficiency in long-time simulation/forecasting due to explicit training of slow time-scale dynamics, and (iv) a flexible framework that is parallelizable and may be integrated with standard numerical time-stepping algorithms. The method is demonstrated on a wide range of nonlinear dynamical systems, including the Van der Pol oscillator, the Lorenz system, the Kuramoto-Sivashinsky equation, and fluid flow pass a cylinder; audio and video signals are also explored. On the sequence generation examples, we benchmark our algorithm against state-of-the-art methods, such as LSTM, reservoir computing, and clockwork RNN. Despite the structural simplicity of our method, it outperforms competing methods on numerical integration.
Recent work has shown that deep learning models can be used to classify land-use data from geospatial satellite imagery. We show that when these deep learning models are trained on data from specific continents/seasons, there is a high degree of variability in model performance on out-of-sample continents/seasons. This suggests that just because a model accurately predicts land-use classes in one continent or season does not mean that the model will accurately predict land-use classes in a different continent or season. We then use clustering techniques on satellite imagery from different continents to visualize the differences in landscapes that make geospatial generalization particularly difficult, and summarize our takeaways for future satellite imagery-related applications.
In recent years the scientific community has devoted much effort in the development of deep learning models for the generation of new molecules with desirable properties (i.e. drugs). This has produced many proposals in literature. However, a systematic comparison among the different VAE methods is still missing. For this reason, we propose an extensive testbed for the evaluation of generative models for drug discovery, and we present the results obtained by many of the models proposed in literature.
Speech enhancement and speech separation are two related tasks, whose purpose is to extract either one or more target speech signals, respectively, from a mixture of sounds generated by several sources. Traditionally, these tasks have been tackled using signal processing and machine learning techniques applied to the available acoustic signals. Since the visual aspect of speech is essentially unaffected by the acoustic environment, visual information from the target speakers, such as lip movements and facial expressions, has also been used for speech enhancement and speech separation systems. In order to efficiently fuse acoustic and visual information, researchers have exploited the flexibility of data-driven approaches, specifically deep learning, achieving strong performance. The ceaseless proposal of a large number of techniques to extract features and fuse multimodal information has highlighted the need for an overview that comprehensively describes and discusses audio-visual speech enhancement and separation based on deep learning. In this paper, we provide a systematic survey of this research topic, focusing on the main elements that characterise the systems in the literature: acoustic features; visual features; deep learning methods; fusion techniques; training targets and objective functions. In addition, we review deep-learning-based methods for speech reconstruction from silent videos and audio-visual sound source separation for non-speech signals, since these methods can be more or less directly applied to audio-visual speech enhancement and separation. Finally, we survey commonly employed audio-visual speech datasets, given their central role in the development of data-driven approaches, and evaluation methods, because they are generally used to compare different systems and determine their performance.
This study presents a deep learning-based speech signal-processing mobile application known as CITISEN. The CITISEN provides three functions: speech enhancement (SE), model adaptation (MA), and background noise conversion (BNC), allowing CITISEN to be used as a platform for utilizing and evaluating SE models and flexibly extend the models to address various noise environments and users. For SE, a pretrained SE model downloaded from the cloud server is used to effectively reduce noise components from instant or saved recordings provided by users. For encountering unseen noise or speaker environments, the MA function is applied to promote CITISEN. A few audio samples recording on a noisy environment are uploaded and used to adapt the pretrained SE model on the server. Finally, for BNC, CITISEN first removes the background noises through an SE model and then mixes the processed speech with new background noise. The novel BNC function can evaluate SE performance under specific conditions, cover people's tracks, and provide entertainment. The experimental results confirmed the effectiveness of SE, MA, and BNC functions. Compared with the noisy speech signals, the enhanced speech signals achieved about 6\% and 33\% of improvements, respectively, in terms of short-time objective intelligibility (STOI) and perceptual evaluation of speech quality (PESQ). With MA, the STOI and PESQ could be further improved by approximately 6\% and 11\%, respectively. Finally, the BNC experiment results indicated that the speech signals converted from noisy and silent backgrounds have a close scene identification accuracy and similar embeddings in an acoustic scene classification model. Therefore, the proposed BNC can effectively convert the background noise of a speech signal and be a data augmentation method when clean speech signals are unavailable.
X-ray diffraction based microscopy techniques such as High Energy Diffraction Microscopy rely on knowledge of the position of diffraction peaks with high precision. These positions are typically computed by fitting the observed intensities in area detector data to a theoretical peak shape such as pseudo-Voigt. As experiments become more complex and detector technologies evolve, the computational cost of such peak detection and shape fitting becomes the biggest hurdle to the rapid analysis required for real-time feedback during in-situ experiments. To this end, we propose BraggNN, a deep learning-based method that can determine peak positions much more rapidly than conventional pseudo-Voigt peak fitting. When applied to a test dataset, BraggNN gives errors of less than 0.29 and 0.57 pixels, relative to the conventional method, for 75% and 95% of the peaks, respectively. When applied to a real experimental dataset, a 3D reconstruction that used peak positions computed by BraggNN yields 15% better results on average as compared to a reconstruction obtained using peak positions determined using conventional 2D pseudo-Voigt fitting. Recent advances in deep learning method implementations and special-purpose model inference accelerators allow BraggNN to deliver enormous performance improvements relative to the conventional method, running, for example, more than 200 times faster than a conventional method on a consumer-class GPU card with out-of-the-box software.
Training a state-of-the-art deep neural network (DNN) is a computationally-expensive and time-consuming process, which incentivizes deep learning developers to debug their DNNs for computational performance. However, effectively performing this debugging requires intimate knowledge about the underlying software and hardware systems---something that the typical deep learning developer may not have. To help bridge this gap, we present Skyline: a new interactive tool for DNN training that supports in-editor computational performance profiling, visualization, and debugging. Skyline's key contribution is that it leverages special computational properties of DNN training to provide (i) interactive performance predictions and visualizations, and (ii) directly manipulatable visualizations that, when dragged, mutate the batch size in the code. As an in-editor tool, Skyline allows users to leverage these diagnostic features to debug the performance of their DNNs during development. An exploratory qualitative user study of Skyline produced promising results; all the participants found Skyline to be useful and easy to use.
Cancer patients have a higher risk of cardiovascular disease (CVD) mortality than the general population. Low dose computed tomography (LDCT) for lung cancer screening offers an opportunity for simultaneous CVD risk estimation in at-risk patients. Our deep learning CVD risk prediction model, trained with 30,286 LDCTs from the National Lung Cancer Screening Trial, achieved an area under the curve (AUC) of 0.871 on a separate test set of 2,085 subjects and identified patients with high CVD mortality risks (AUC of 0.768). We validated our model against ECG-gated cardiac CT based markers, including coronary artery calcification (CAC) score, CAD-RADS score, and MESA 10-year risk score from an independent dataset of 335 subjects. Our work shows that, in high-risk patients, deep learning can convert LDCT for lung cancer screening into a dual-screening quantitative tool for CVD risk estimation.
Most of the automatic fire alarm systems detect the fire presence through sensors like thermal, smoke, or flame. One of the new approaches to the problem is the use of images to perform the detection. The image approach is promising since it does not need specific sensors and can be easily embedded in different devices. However, besides the high performance, the computational cost of the used deep learning methods is a challenge to their deployment in portable devices. In this work, we propose a new deep learning architecture that requires fewer floating-point operations (flops) for fire recognition. Additionally, we propose a portable approach for fire recognition and the use of modern techniques such as inverted residual block, convolutions like depth-wise, and octave, to reduce the model's computational cost. The experiments show that our model keeps high accuracy while substantially reducing the number of parameters and flops. One of our models presents 71\% fewer parameters than FireNet, while still presenting competitive accuracy and AUROC performance. The proposed methods are evaluated on FireNet and FiSmo datasets. The obtained results are promising for the implementation of the model in a mobile device, considering the reduced number of flops and parameters acquired.
Since 2012, Deep learning has revolutionized Artificial Intelligence and has achieved state-of-the-art outcomes in different domains, ranging from Image Classification to Speech Generation. Though it has many potentials, our current architectures come with the pre-requisite of large amounts of data. Few-Shot Learning (also known as one-shot learning) is a sub-field of machine learning that aims to create such models that can learn the desired objective with less data, similar to how humans learn. In this paper, we have reviewed some of the well-known deep learning-based approaches towards few-shot learning. We have discussed the recent achievements, challenges, and possibilities of improvement of few-shot learning based deep learning architectures. Our aim for this paper is threefold: (i) Give a brief introduction to deep learning architectures for few-shot learning with pointers to core references. (ii) Indicate how deep learning has been applied to the low-data regime, from data preparation to model training. and, (iii) Provide a starting point for people interested in experimenting and perhaps contributing to the field of few-shot learning by pointing out some useful resources and open-source code. Our code is available at Github: https://github.com/shruti-jadon/Hands-on-One-Shot-Learning.
How does a user's prior experience with deep learning impact accuracy? We present an initial study based on 31 participants with different levels of experience. Their task is to perform hyperparameter optimization for a given deep learning architecture. The results show a strong positive correlation between the participant's experience and the final performance. They additionally indicate that an experienced participant finds better solutions using fewer resources on average. The data suggests furthermore that participants with no prior experience follow random strategies in their pursuit of optimal hyperparameters. Our study investigates the subjective human factor in comparisons of state of the art results and scientific reproducibility in deep learning.