Designing RNA molecules has garnered recent interest in medicine, synthetic biology, biotechnology and bioinformatics since many functional RNA molecules were shown to be involved in regulatory processes for transcription, epigenetics and translation. Since an RNA's function depends on its structural properties, the RNA Design problem is to find an RNA sequence which satisfies given structural constraints. Here, we propose a new algorithm for the RNA Design problem, dubbed LEARNA. LEARNA uses deep reinforcement learning to train a policy network to sequentially design an entire RNA sequence given a specified target structure. By meta-learning across 65000 different RNA Design tasks for one hour on 20 CPU cores, our extension Meta-LEARNA constructs an RNA Design policy that can be applied out of the box to solve novel RNA Design tasks. Methodologically, for what we believe to be the first time, we jointly optimize over a rich space of architectures for the policy network, the hyperparameters of the training procedure and the formulation of the decision process. Comprehensive empirical results on two widely-used RNA Design benchmarks, as well as a third one that we introduce, show that our approach achieves new state-of-the-art performance on the former while also being orders of magnitudes faster in reaching the previous state-of-the-art performance. In an ablation study, we analyze the importance of our method's different components.
This paper proposes an efficient neural network (NN) architecture design methodology called Chameleon that honors given resource constraints. Instead of developing new building blocks or using computationally-intensive reinforcement learning algorithms, our approach leverages existing efficient network building blocks and focuses on exploiting hardware traits and adapting computation resources to fit target latency and/or energy constraints. We formulate platform-aware NN architecture search in an optimization framework and propose a novel algorithm to search for optimal architectures aided by efficient accuracy and resource (latency and/or energy) predictors. At the core of our algorithm lies an accuracy predictor built atop Gaussian Process with Bayesian optimization for iterative sampling. With a one-time building cost for the predictors, our algorithm produces state-of-the-art model architectures on different platforms under given constraints in just minutes. Our results show that adapting computation resources to building blocks is critical to model performance. Without the addition of any bells and whistles, our models achieve significant accuracy improvements against state-of-the-art hand-crafted and automatically designed architectures. We achieve 73.8% and 75.3% top-1 accuracy on ImageNet at 20ms latency on a mobile CPU and DSP. At reduced latency, our models achieve up to 8.5% (4.8%) and 6.6% (9.3%) absolute top-1 accuracy improvements compared to MobileNetV2 and MnasNet, respectively, on a mobile CPU (DSP), and 2.7% (4.6%) and 5.6% (2.6%) accuracy gains over ResNet-101 and ResNet-152, respectively, on an Nvidia GPU (Intel CPU).
The main goal in many fields in the empirical sciences is to discover causal relationships among a set of variables from observational data. PC algorithm is one of the promising solutions to learn underlying causal structure by performing a number of conditional independence tests. In this paper, we propose a novel GPU-based parallel algorithm, called cuPC, to execute an order-independent version of PC. The proposed solution has two variants, cuPC-E and cuPC-S, which parallelize PC in two different ways for multivariate normal distribution. Experimental results show the scalability of the proposed algorithms with respect to the number of variables, the number of samples, and different graph densities. For instance, in one of the most challenging datasets, the runtime is reduced from more than 11 hours to about 4 seconds. On average, cuPC-E and cuPC-S achieve 500 X and 1300 X speedup, respectively, compared to serial implementation on CPU. The source code of cuPC is available online [1].
Time series anomaly detection is usually formulated as finding outlier data points relative to some usual data, which is also an important problem in industry and academia. To ensure systems working stably, internet companies, banks and other companies need to monitor time series, which is called KPI (Key Performance Indicators), such as CPU used, number of orders, number of online users and so on. However, millions of time series have several shapes (e.g. seasonal KPIs, KPIs of timed tasks and KPIs of CPU used), so that it is very difficult to use a simple statistical model to detect anomaly for all kinds of time series. Although some anomaly detectors have developed many years and some supervised models are also available in this field, we find many methods have their own disadvantages. In this paper, we present our system, which is based on deep feedforward neural network and detect anomaly points of time series. The main difference between our system and other systems based on supervised models is that we do not need feature engineering of time series to train deep feedforward neural network in our system, which is essentially an end-to-end system.
This study proposes a logic architecture for the high-speed and power efficiently training of a gradient boosting decision tree model of binary classification. We implemented the proposed logic architecture on an FPGA and compared training time and power efficiency with three general GBDT software libraries using CPU and GPU. The training speed of the logic architecture on the FPGA was 26-259 times faster than the software libraries. The power efficiency of the logic architecture was 90-1,104 times higher than the software libraries. The results show that the logic architecture suits for high-performance and edge computing.
We propose a new multi-instance dynamic RGB-D SLAM system using an object-level octree-based volumetric representation. It can provide robust camera tracking in dynamic environments and at the same time, continuously estimate geometric, semantic, and motion properties for arbitrary objects in the scene. For each incoming frame, we perform instance segmentation to detect objects and refine mask boundaries using geometric and motion information. Meanwhile, we estimate the pose of each existing moving object using an object-oriented tracking method and robustly track the camera pose against the static scene. Based on the estimated camera pose and object poses, we associate segmented masks with existing models and incrementally fuse corresponding colour, depth, semantic, and foreground object probabilities into each object model. In contrast to existing approaches, our system is the first system to generate an object-level dynamic volumetric map from a single RGB-D camera, which can be used directly for robotic tasks. Our method can run at 2-3 Hz on a CPU, excluding the instance segmentation part. We demonstrate its effectiveness by quantitatively and qualitatively testing it on both synthetic and real-world sequences.
Deep neural network models used for medical image segmentation are large because they are trained with high-resolution three-dimensional (3D) images. Graphics processing units (GPUs) are widely used to accelerate the trainings. However, the memory on a GPU is not large enough to train the models. A popular approach to tackling this problem is patch-based method, which divides a large image into small patches and trains the models with these small patches. However, this method would degrade the segmentation quality if a target object spans multiple patches. In this paper, we propose a novel approach for 3D medical image segmentation that utilizes the data-swapping, which swaps out intermediate data from GPU memory to CPU memory to enlarge the effective GPU memory size, for training high-resolution 3D medical images without patching. We carefully tuned parameters in the data-swapping method to obtain the best training performance for 3D U-Net, a widely used deep neural network model for medical image segmentation. We applied our tuning to train 3D U-Net with full-size images of 192 x 192 x 192 voxels in brain tumor dataset. As a result, communication overhead, which is the most important issue, was reduced by 17.1%. Compared with the patch-based method for patches of 128 x 128 x 128 voxels, our training for full-size images achieved improvement on the mean Dice score by 4.48% and 5.32 % for detecting whole tumor sub-region and tumor core sub-region, respectively. The total training time was reduced from 164 hours to 47 hours, resulting in 3.53 times of acceleration.
We present a highly parallel method for accurate and efficient variational deformable 3D image registration on a consumer-grade graphics processing unit (GPU). We build on recent matrix-free variational approaches and specialize the concepts to the massively-parallel manycore architecture provided by the GPU. Compared to a parallel and optimized CPU implementation, this allows us to achieve an average speedup of 32.53 on 986 real-world CT thorax-abdomen follow-up scans. At a resolution of approximately $256^3$ voxels, the average runtime is 1.99 seconds for the full registration. On the publicly available DIR-lab benchmark, our method ranks third with respect to average landmark error at an average runtime of 0.32 seconds.
High efficiency in precision farming depends on accurate tools to perform weed detection and mapping of crops. This allows for precise removal of harmful weeds with a lower amount of pesticides, as well as increase of the harvest's yield by providing the farmer with valuable information. In this paper, we address the problem of fully automatic stem detection from image data for this purpose. Our approach runs on mobile agricultural robots taking RGB images. After processing the images to obtain a vegetation mask, our approach separates each plant into its individual leaves and later estimates a precise stem position. This allows an upstream mapping algorithm to add the high-resolution stem positions as a semantic aggregate to the global map of the robot, which can be used for weeding and for analyzing crop statistics. We implemented our approach and thoroughly tested it on three different datasets with vegetation masks and stem position ground truth. The experiments presented in this paper conclude that our module is able to detect leaves and estimate the stem's position at a rate of 56 Hz on a single CPU. We furthermore provide the software to the community.
Purpose: To perform and evaluate water-fat signal separation of whole-body gradient echo scans using convolutional neural networks. Methods: Whole-body gradient echo scans of 240 subjects, each consisting of 5 bipolar echoes, were used. Reference fat fraction maps were created using a conventional method. Convolutional neural networks, more specifically 2D U-nets, were trained using 5-fold cross-validation with 1 or several echoes as input, using the squared difference between the output and the reference fat fraction maps as the loss function. The outputs of the networks were assessed by the loss function, measured liver fat fractions, and visually. Training was performed using a graphics processing unit (GPU). Inference was performed using the GPU as well as a central processing unit (CPU). Results: The loss curves indicated convergence, and the final loss of the validation data decreased when using more echoes as input. The liver fat fractions could be estimated using only 1 echo, but results were improved by use of more echoes. Visual assessment found the quality of the outputs of the networks to be similar to the reference even when using only 1 echo, with slight improvements when using more echoes. Training a network took at most 28.6 h. Inference time of a whole-body scan took at most 3.7 s using the GPU and 5.8 min using the CPU. Conclusion: It is possible to perform water-fat signal separation of whole-body gradient echo scans using convolutional neural networks. Separation was possible using only 1 echo, although using more echoes improved the results.
More accurate machine learning models often demand more computation and memory at test time, making them difficult to deploy on CPU- or memory-constrained devices. Teacher-student compression (TSC), also known as distillation, alleviates this burden by training a less expensive student model to mimic the expensive teacher model while maintaining most of the original accuracy. However, when fresh data is unavailable for the compression task, the teacher's training data is typically reused, leading to suboptimal compression. In this work, we propose to augment the compression dataset with synthetic data from a generative adversarial network (GAN) designed to approximate the training data distribution. Our GAN-assisted TSC (GAN-TSC) significantly improves student accuracy for expensive models such as large random forests and deep neural networks on both tabular and image datasets. Building on these results, we propose a comprehensive metric---the TSC Score---to evaluate the quality of synthetic datasets based on their induced TSC performance. The TSC Score captures both data diversity and class affinity, and we illustrate its benefits over the popular Inception Score in the context of image classification.
The Earth Mover's Distance (EMD) is a state-of-the art metric for comparing discrete probability distributions, but its high distinguishability comes at a high cost in computational complexity. Even though linear-complexity approximation algorithms have been proposed to improve its scalability, these algorithms are either limited to vector spaces with only a few dimensions or they become ineffective when the degree of overlap between the probability distributions is high. We propose novel approximation algorithms that overcome both of these limitations, yet still achieve linear time complexity. All our algorithms are data parallel, and thus, we take advantage of massively parallel computing engines, such as Graphics Processing Units (GPUs). On the popular text-based 20 Newsgroups dataset, the new algorithms are four orders of magnitude faster than a multi-threaded CPU implementation of Word Mover's Distance and match its nearest-neighbors-search accuracy. On MNIST images, the new algorithms are four orders of magnitude faster than a GPU implementation of the Sinkhorn's algorithm while offering a slightly higher nearest-neighbors-search accuracy.
Using Intel's Loihi neuromorphic research chip and ABR's Nengo Deep Learning toolkit, we analyze the inference speed, dynamic power consumption, and energy cost per inference of a two-layer neural network keyword spotter trained to recognize a single phrase. We perform comparative analyses of this keyword spotter running on more conventional hardware devices including a CPU, a GPU, Nvidia's Jetson TX1, and the Movidius Neural Compute Stick. Our results indicate that for this inference application, Loihi outperforms all of these alternatives on an energy cost per inference basis while maintaining equivalent inference accuracy. Furthermore, an analysis of tradeoffs between network size, inference speed, and energy cost indicates that Loihi's comparative advantage over other low-power computing devices improves for larger networks.
TensorFlow is a popular deep learning framework used by data scientists to solve a wide-range of machine learning and deep learning problems such as image classification and speech recognition. It also operates at a large scale and in heterogeneous environments --- it allows users to train neural network models or deploy them for inference using GPUs, CPUs and deep learning specific custom-designed hardware such as TPUs. Even though TensorFlow supports a variety of optimized backends, realizing the best performance using a backend may require additional efforts. For instance, getting the best performance from a CPU backend requires careful tuning of its threading model. Unfortunately, the best tuning approach used today is manual, tedious, time-consuming, and, more importantly, may not guarantee the best performance. In this paper, we develop an automatic approach, called TensorTuner, to search for optimal parameter settings of TensorFlow's threading model for CPU backends. We evaluate TensorTuner on both Eigen and Intel's MKL CPU backends using a set of neural networks from TensorFlow's benchmarking suite. Our evaluation results demonstrate that the parameter settings found by TensorTuner produce 2% to 123% performance improvement for the Eigen CPU backend and 1.5% to 28% performance improvement for the MKL CPU backend over the performance obtained using their best-known parameter settings. This highlights the fact that the default parameter settings in Eigen CPU backend are not the ideal settings; and even for a carefully hand-tuned MKL backend, the settings may be sub-optimal. Our evaluations also revealed that TensorTuner is efficient at finding the optimal settings --- it is able to converge to the optimal settings quickly by pruning more than 90% of the parameter search space.
Speech, Music and Noise classification/segmentation is an important preprocessing step for audio processing/indexing. To this end, we propose a novel 1D Convolutional Neural Network (CNN) - SwishNet. It is a fast and lightweight architecture that operates on MFCC features which is suitable to be added to the front-end of an audio processing pipeline. We showed that the performance of our network can be improved by distilling knowledge from a 2D CNN, pretrained on ImageNet. We investigated the performance of our network on the MUSAN corpus - an openly available comprehensive collection of noise, music and speech samples, suitable for deep learning. The proposed network achieved high overall accuracy in clip (length of 0.5-2s) classification (>97% accuracy) and frame-wise segmentation (>93% accuracy) tasks with even higher accuracy (>99%) in speech/non-speech discrimination task. To verify the robustness of our model, we trained it on MUSAN and evaluated it on a different corpus - GTZAN and found good accuracy with very little fine-tuning. We also demonstrated that our model is fast on both CPU and GPU, consumes a low amount of memory and is suitable for implementation in embedded systems.
A new method is proposed for removing text from natural images. The challenge is to first accurately localize text on the stroke-level and then replace it with a visually plausible background. Unlike previous methods that require image patches to erase scene text, our method, namely ensconce network (EnsNet), can operate end-to-end on a single image without any prior knowledge. The overall structure is an end-to-end trainable FCN-ResNet-18 network with a conditional generative adversarial network (cGAN). The feature of the former is first enhanced by a novel lateral connection structure and then refined by four carefully designed losses: multiscale regression loss and content loss, which capture the global discrepancy of different level features; texture loss and total variation loss, which primarily target filling the text region and preserving the reality of the background. The latter is a novel local-sensitive GAN, which attentively assesses the local consistency of the text erased regions. Both qualitative and quantitative sensitivity experiments on synthetic images and the ICDAR 2013 dataset demonstrate that each component of the EnsNet is essential to achieve a good performance. Moreover, our EnsNet can significantly outperform previous state-of-the-art methods in terms of all metrics. In addition, a qualitative experiment conducted on the SMBNet dataset further demonstrates that the proposed method can also preform well on general object (such as pedestrians) removal tasks. EnsNet is extremely fast, which can preform at 333 fps on an i5-8600 CPU device.
Network compression reduces the computational complexity and memory consumption of deep neural networks by reducing the number of parameters. In SVD-based network compression, the right rank needs to be decided for every layer of the network. In this paper, we propose an efficient method for obtaining the rank configuration of the whole network. Unlike previous methods which consider each layer separately, our method considers the whole network to choose the right rank configuration. We propose novel accuracy metrics to represent the accuracy and complexity relationship for a given neural network. We use these metrics in a non-iterative fashion to obtain the right rank configuration which satisfies the constraints on FLOPs and memory while maintaining sufficient accuracy. Experiments show that our method provides better compromise between accuracy and computational complexity/memory consumption while performing compression at much higher speed. For VGG-16 our network can reduce the FLOPs by 25% and improve accuracy by 0.7% compared to the baseline, while requiring only 3 minutes on a CPU to search for the right rank configuration. Previously, similar results were achieved in 4 hours with 8 GPUs. The proposed method can be used for lossless compression of a neural network as well. The better accuracy and complexity compromise, as well as the extremely fast speed of our method makes it suitable for neural network compression.
In this work we adapt multi-person pose estimation architecture to use it on edge devices. We follow the bottom-up approach from OpenPose, the winner of COCO 2016 Keypoints Challenge, because of its decent quality and robustness to number of people inside the frame. With proposed network design and optimized post-processing code the full solution runs at 28 frames per second (fps) on Intel$\unicode{xAE}$ NUC 6i7KYB mini PC and 26 fps on Core$^{TM}$ i7-6850K CPU. The network model has 4.1M parameters and 9 billions floating-point operations (GFLOPs) complexity, which is just ~15% of the baseline 2-stage OpenPose with almost the same quality. The code and model are available as a part of Intel$\unicode{xAE}$ OpenVINO$^{TM}$ Toolkit.
GPUs have limited memory and it is difficult to train wide and/or deep models that cause the training process to go out of memory. It is shown in this paper how an open source tool called Large Model Support (LMS) can utilize a high bandwidth NVLink connection between CPUs and GPUs to accomplish training of deep convolutional networks. LMS performs tensor swapping between CPU memory and GPU memory such that only a minimal number of tensors required in a training step are kept in the GPU memory. It is also shown how LMS can be combined with an MPI based distributed deep learning module to train models in a data-parallel fashion across multiple GPUs, such that each GPU is utilizing the CPU memory for tensor swapping. The hardware architecture that enables the high bandwidth GPU link with the CPU is discussed as well as the associated set of software tools that are available as the PowerAI package.
Fully-convolutional neural networks have achieved superior performance in a variety of image segmentation tasks. However, their training requires laborious manual annotation of large datasets, as well as acceleration by parallel processors with high-bandwidth memory, such as GPUs. We show that simple models can achieve competitive accuracy for organ segmentation on CT images when trained with extensive data augmentation, which leverages existing graphics hardware to quickly apply geometric and photometric transformations to 3D image data. On 3 mm^3 CT volumes, our GPU implementation is 2.6-8X faster than a widely-used CPU version, including communication overhead. We also show how to automatically generate training labels using rudimentary morphological operations, which are efficiently computed by 3D Fourier transforms. We combined fully-automatic labels for the lungs and bone with semi-automatic ones for the liver, kidneys and bladder, to create a dataset of 130 labeled CT scans. To achieve the best results from data augmentation, our model uses the intersection-over-union (IOU) loss function, a close relative of the Dice loss. We discuss its mathematical properties and explain why it outperforms the usual weighted cross-entropy loss for unbalanced segmentation tasks. We conclude that there is no unique IOU loss function, as the naive one belongs to a broad family of functions with the same essential properties. When combining data augmentation with the IOU loss, our model achieves a Dice score of 78-92% for each organ. The trained model, code and dataset will be made publicly available, to further medical imaging research.