PyTorchKR Papers

최신 논문

총 5,575편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1509.09308

Fast Algorithms for Convolutional Neural Networks

Deep convolutional neural networks take GPU days of compute time to train on large data sets. Pedestrian detection for self driving cars requires very low latency. Image recognition for mobile phones is constrained by limited processing resources. The success of convolutional neural networks in these situations is limited by how fast we can compute them. Conventional FFT based convolution is fast for large filters, but state of the art convolutional neural networks use small, 3x3 filters. We introduce a new class of fast algorithms for convolutional neural networks using Winograd's minimal filtering algorithms. The algorithms compute minimal complexity convolution over small tiles, which makes them fast with small filters and small batch sizes. We benchmark a GPU implementation of our algorithm with the VGG network and show state of the art throughput at batch sizes from 1 to 64.

arXiv:1511.07106

Multi-Volume High Resolution RGB-D Mapping with Dynamic Volume Placement

We present a novel RGB-D mapping system for generating 3D maps over spatially extended regions with higher resolution than current methods using multiple, dynamically placed mapping volumes. Our method takes in RGB-D frames and dynamically assigns multiple mapping volumes to the environment, exchanging mapping volumes between the CPU and GPU. Mapping volumes are added or removed as needed to allow for spatially extended, high resolution mapping. Our system is designed to maximize the resolution possible for such volumetric methods, while working on an unbounded space.

arXiv:1508.01292

Compact Convolutional Neural Network Cascade for Face Detection

The problem of faces detection in images or video streams is a classical problem of computer vision. The multiple solutions of this problem have been proposed, but the question of their optimality is still open. Many algorithms achieve a high quality face detection, but at the cost of high computational complexity. This restricts their application in the real-time systems. This paper presents a new solution of the frontal face detection problem based on compact convolutional neural networks cascade. The test results on FDDB dataset show that it is competitive with state-of-the-art algorithms. This proposed detector is implemented using three technologies: SSE/AVX/AVX2 instruction sets for Intel CPUs, Nvidia CUDA, OpenCL. The detection speed of our approach considerably exceeds all the existing CPU-based and GPU-based algorithms. Because of high computational efficiency, our detector can processing 4K Ultra HD video stream in real time (up to 27 fps) on mobile platforms (Intel Ivy Bridge CPUs and Nvidia Kepler GPUs) in searching objects with the dimension 60x60 pixels or higher. At the same time its performance weakly dependent on the background and number of objects in scene. This is achieved by the asynchronous computation of stages in the cascade.

arXiv:1512.01641

Unsupervised comparable corpora preparation and exploration for bi-lingual translation equivalents

The multilingual nature of the world makes translation a crucial requirement today. Parallel dictionaries constructed by humans are a widely-available resource, but they are limited and do not provide enough coverage for good quality translation purposes, due to out-of-vocabulary words and neologisms. This motivates the use of statistical translation systems, which are unfortunately dependent on the quantity and quality of training data. Such systems have a very limited availability especially for some languages and very narrow text domains. In this research we present our improvements to current comparable corpora mining methodologies by re- implementation of the comparison algorithms (using Needleman-Wunch algorithm), introduction of a tuning script and computation time improvement by GPU acceleration. Experiments are carried out on bilingual data extracted from the Wikipedia, on various domains. For the Wikipedia itself, additional cross-lingual comparison heuristics were introduced. The modifications made a positive impact on the quality and quantity of mined data and on the translation quality.

arXiv:1512.01274

MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems

MXNet is a multi-language machine learning (ML) library to ease the development of ML algorithms, especially for deep neural networks. Embedded in the host language, it blends declarative symbolic expression with imperative tensor computation. It offers auto differentiation to derive gradients. MXNet is computation and memory efficient and runs on various heterogeneous systems, ranging from mobile devices to distributed GPU clusters. This paper describes both the API design and the system implementation of MXNet, and explains how embedding of both symbolic expression and tensor operation is handled in a unified fashion. Our preliminary experiments reveal promising results on large scale deep neural network applications using multiple GPU machines.

arXiv:1509.06004

A Parallel Framework for Parametric Maximum Flow Problems in Image Segmentation

This paper presents a framework that supports the implementation of parallel solutions for the widespread parametric maximum flow computational routines used in image segmentation algorithms. The framework is based on supergraphs, a special construction combining several image graphs into a larger one, and works on various architectures (multi-core or GPU), either locally or remotely in a cluster of computing nodes. The framework can also be used for performance evaluation of parallel implementations of maximum flow algorithms. We present the case study of a state-of-the-art image segmentation algorithm based on graph cuts, Constrained Parametric Min-Cut (CPMC), that uses the parallel framework to solve parametric maximum flow problems, based on a GPU implementation of the well-known push-relabel algorithm. Our results indicate that real-time implementations based on the proposed techniques are possible.

arXiv:1512.03375

Convolutional Monte Carlo Rollouts in Go

In this work, we present a MCTS-based Go-playing program which uses convolutional networks in all parts. Our method performs MCTS in batches, explores the Monte Carlo search tree using Thompson sampling and a convolutional network, and evaluates convnet-based rollouts on the GPU. We achieve strong win rates against open source Go programs and attain competitive results against state of the art convolutional net-based Go-playing programs.

arXiv:1512.06216

Poseidon: A System Architecture for Efficient GPU-based Deep Learning on Multiple Machines

Deep learning (DL) has achieved notable successes in many machine learning tasks. A number of frameworks have been developed to expedite the process of designing and training deep neural networks (DNNs), such as Caffe, Torch and Theano. Currently they can harness multiple GPUs on a single machine, but are unable to use GPUs that are distributed across multiple machines; as even average-sized DNNs can take days to train on a single GPU with 100s of GBs to TBs of data, distributed GPUs present a prime opportunity for scaling up DL. However, the limited bandwidth available on commodity Ethernet networks presents a bottleneck to distributed GPU training, and prevents its trivial realization. To investigate how to adapt existing frameworks to efficiently support distributed GPUs, we propose Poseidon, a scalable system architecture for distributed inter-machine communication in existing DL frameworks. We integrate Poseidon with Caffe and evaluate its performance at training DNNs for object recognition. Poseidon features three key contributions that accelerate DNN training on clusters: (1) a three-level hybrid architecture that allows Poseidon to support both CPU-only and GPU-equipped clusters, (2) a distributed wait-free backpropagation (DWBP) algorithm to improve GPU utilization and to balance communication, and (3) a structure-aware communication protocol (SACP) to minimize communication overheads. We empirically show that Poseidon converges to same objectives as a single machine, and achieves state-of-art training speedup across multiple models and well-established datasets using a commodity GPU cluster of 8 nodes (e.g. 4.5x speedup on AlexNet, 4x on GoogLeNet, 4x on CIFAR-10). On the much larger ImageNet22K dataset, Poseidon with 8 nodes achieves better speedup and competitive accuracy to recent CPU-based distributed systems such as Adam and Le et al., which use 10s to 1000s of nodes.

arXiv:1506.01497공식 코드 · Papers with Code

Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks

State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet and Fast R-CNN have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN into a single network by sharing their convolutional features---using the recently popular terminology of neural networks with 'attention' mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model, our detection system has a frame rate of 5fps (including all steps) on a GPU, while achieving state-of-the-art object detection accuracy on PASCAL VOC 2007, 2012, and MS COCO datasets with only 300 proposals per image. In ILSVRC and COCO 2015 competitions, Faster R-CNN and RPN are the foundations of the 1st-place winning entries in several tracks. Code has been made publicly available.

arXiv:1408.3264

A brief survey on deep belief networks and introducing a new object oriented toolbox (DeeBNet)

Nowadays, this is very popular to use the deep architectures in machine learning. Deep Belief Networks (DBNs) are deep architectures that use stack of Restricted Boltzmann Machines (RBM) to create a powerful generative model using training data. DBNs have many ability like feature extraction and classification that are used in many applications like image processing, speech processing and etc. This paper introduces a new object oriented MATLAB toolbox with most of abilities needed for the implementation of DBNs. In the new version, the toolbox can be used in Octave. According to the results of the experiments conducted on MNIST (image), ISOLET (speech), and 20 Newsgroups (text) datasets, it was shown that the toolbox can learn automatically a good representation of the input from unlabeled data with better discrimination between different classes. Also on all datasets, the obtained classification errors are comparable to those of state of the art classifiers. In addition, the toolbox supports different sampling methods (e.g. Gibbs, CD, PCD and our new FEPCD method), different sparsity methods (quadratic, rate distortion and our new normal method), different RBM types (generative and discriminative), using GPU, etc. The toolbox is a user-friendly open source software and is freely available on the website http://ceit.aut.ac.ir/~keyvanrad/DeeBNet%20Toolbox.html .

arXiv:1511.00175

FireCaffe: near-linear acceleration of deep neural network training on compute clusters

Long training times for high-accuracy deep neural networks (DNNs) impede research into new DNN architectures and slow the development of high-accuracy DNNs. In this paper we present FireCaffe, which successfully scales deep neural network training across a cluster of GPUs. We also present a number of best practices to aid in comparing advancements in methods for scaling and accelerating the training of deep neural networks. The speed and scalability of distributed algorithms is almost always limited by the overhead of communicating between servers; DNN training is not an exception to this rule. Therefore, the key consideration here is to reduce communication overhead wherever possible, while not degrading the accuracy of the DNN models that we train. Our approach has three key pillars. First, we select network hardware that achieves high bandwidth between GPU servers -- Infiniband or Cray interconnects are ideal for this. Second, we consider a number of communication algorithms, and we find that reduction trees are more efficient and scalable than the traditional parameter server approach. Third, we optionally increase the batch size to reduce the total quantity of communication during DNN training, and we identify hyperparameters that allow us to reproduce the small-batch accuracy while training with large batch sizes. When training GoogLeNet and Network-in-Network on ImageNet, we achieve a 47x and 39x speedup, respectively, when training on a cluster of 128 GPUs.

arXiv:1601.05030공식 코드 · Papers with Code

PN-Net: Conjoined Triple Deep Network for Learning Local Image Descriptors

In this paper we propose a new approach for learning local descriptors for matching image patches. It has recently been demonstrated that descriptors based on convolutional neural networks (CNN) can significantly improve the matching performance. Unfortunately their computational complexity is prohibitive for any practical application. We address this problem and propose a CNN based descriptor with improved matching performance, significantly reduced training and execution time, as well as low dimensionality. We propose to train the network with triplets of patches that include a positive and negative pairs. To that end we introduce a new loss function that exploits the relations within the triplets. We compare our approach to recently introduced MatchNet and DeepCompare and demonstrate the advantages of our descriptor in terms of performance, memory footprint and speed i.e. when run in GPU, the extraction time of our 128 dimensional feature is comparable to the fastest available binary descriptors such as BRIEF and ORB.

arXiv:1601.06062

3-D/2-D Registration of Cardiac Structures by 3-D Contrast Agent Distribution Estimation

For augmented fluoroscopy during cardiac catheter ablation procedures, a preoperatively acquired 3-D model of the left atrium of the patient can be registered to X-ray images. Therefore the 3D-model is matched with the contrast agent based appearance of the left atrium. Commonly, only small amounts of contrast agent (CA) are used to locate the left atrium. This is why we focus on robust registration methods that work also if the structure of interest is only partially contrasted. In particular, we propose two similarity measures for CA-based registration: The first similarity measure, explicit apparent edges, focuses on edges of the patient anatomy made visible by contrast agent and can be computed quickly on the GPU. The second novel similarity measure computes a contrast agent distribution estimate (CADE) inside the 3-D model and rates its consistency with the CA seen in biplane fluoroscopic images. As the CADE computation involves a reconstruction of CA in 3-D using the CA within the fluoroscopic images, it is slower. Using a combination of both methods, our evaluation on 11 well-contrasted clinical datasets yielded an error of 7.9+/-6.3 mm over all frames. For 10 datasets with little CA, we obtained an error of 8.8+/-6.7 mm. Our new methods outperform a registration based on the projected shadow significantly (p<0.05).

arXiv:1602.04283

Deep Learning on FPGAs: Past, Present, and Future

The rapid growth of data size and accessibility in recent years has instigated a shift of philosophy in algorithm design for artificial intelligence. Instead of engineering algorithms by hand, the ability to learn composable systems automatically from massive amounts of data has led to ground-breaking performance in important domains such as computer vision, speech recognition, and natural language processing. The most popular class of techniques used in these domains is called deep learning, and is seeing significant attention from industry. However, these models require incredible amounts of data and compute power to train, and are limited by the need for better hardware acceleration to accommodate scaling beyond current data and model sizes. While the current solution has been to use clusters of graphics processing units (GPU) as general purpose processors (GPGPU), the use of field programmable gate arrays (FPGA) provide an interesting alternative. Current trends in design tools for FPGAs have made them more compatible with the high-level software practices typically practiced in the deep learning community, making FPGAs more accessible to those who build and deploy models. Since FPGA architectures are flexible, this could also allow researchers the ability to explore model-level optimizations beyond what is possible on fixed architectures such as GPUs. As well, FPGAs tend to provide high performance per watt of power consumption, which is of particular importance for application scientists interested in large scale server-based deployment or resource-limited embedded applications. This review takes a look at deep learning and FPGAs from a hardware acceleration perspective, identifying trends and innovations that make these technologies a natural fit, and motivates a discussion on how FPGAs may best serve the needs of the deep learning community moving forward.

arXiv:1602.04124

Fast and Robust Hand Tracking Using Detection-Guided Optimization

Markerless tracking of hands and fingers is a promising enabler for human-computer interaction. However, adoption has been limited because of tracking inaccuracies, incomplete coverage of motions, low framerate, complex camera setups, and high computational requirements. In this paper, we present a fast method for accurately tracking rapid and complex articulations of the hand using a single depth camera. Our algorithm uses a novel detection-guided optimization strategy that increases the robustness and speed of pose estimation. In the detection step, a randomized decision forest classifies pixels into parts of the hand. In the optimization step, a novel objective function combines the detected part labels and a Gaussian mixture representation of the depth to estimate a pose that best fits the depth. Our approach needs comparably less computational resources which makes it extremely fast (50 fps without GPU support). The approach also supports varying static, or moving, camera-to-scene arrangements. We show the benefits of our method by evaluating on public datasets and comparing against previous work.

arXiv:1510.00149

Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding

Neural networks are both computationally intensive and memory intensive, making them difficult to deploy on embedded systems with limited hardware resources. To address this limitation, we introduce "deep compression", a three stage pipeline: pruning, trained quantization and Huffman coding, that work together to reduce the storage requirement of neural networks by 35x to 49x without affecting their accuracy. Our method first prunes the network by learning only the important connections. Next, we quantize the weights to enforce weight sharing, finally, we apply Huffman coding. After the first two steps we retrain the network to fine tune the remaining connections and the quantized centroids. Pruning, reduces the number of connections by 9x to 13x; Quantization then reduces the number of bits that represent each connection from 32 to 5. On the ImageNet dataset, our method reduced the storage required by AlexNet by 35x, from 240MB to 6.9MB, without loss of accuracy. Our method reduced the size of VGG-16 by 49x from 552MB to 11.3MB, again with no loss of accuracy. This allows fitting the model into on-chip SRAM cache rather than off-chip DRAM memory. Our compression method also facilitates the use of complex neural networks in mobile applications where application size and download bandwidth are constrained. Benchmarked on CPU, GPU and mobile GPU, compressed network has 3x to 4x layerwise speedup and 3x to 7x better energy efficiency.

arXiv:1602.06645

Creating Simplified 3D Models with High Quality Textures

This paper presents an extension to the KinectFusion algorithm which allows creating simplified 3D models with high quality RGB textures. This is achieved through (i) creating model textures using images from an HD RGB camera that is calibrated with Kinect depth camera, (ii) using a modified scheme to update model textures in an asymmetrical colour volume that contains a higher number of voxels than that of the geometry volume, (iii) simplifying dense polygon mesh model using quadric-based mesh decimation algorithm, and (iv) creating and mapping 2D textures to every polygon in the output 3D model. The proposed method is implemented in real-time by means of GPU parallel processing. Visualization via ray casting of both geometry and colour volumes provides users with a real-time feedback of the currently scanned 3D model. Experimental results show that the proposed method is capable of keeping the model texture quality even for a heavily decimated model and that, when reconstructing small objects, photorealistic RGB textures can still be reconstructed.

arXiv:1511.06530

Compression of Deep Convolutional Neural Networks for Fast and Low Power Mobile Applications

Although the latest high-end smartphone has powerful CPU and GPU, running deeper convolutional neural networks (CNNs) for complex tasks such as ImageNet classification on mobile devices is challenging. To deploy deep CNNs on mobile devices, we present a simple and effective scheme to compress the entire CNN, which we call one-shot whole network compression. The proposed scheme consists of three steps: (1) rank selection with variational Bayesian matrix factorization, (2) Tucker decomposition on kernel tensor, and (3) fine-tuning to recover accumulated loss of accuracy, and each step can be easily implemented using publicly available tools. We demonstrate the effectiveness of the proposed scheme by testing the performance of various compressed CNNs (AlexNet, VGGS, GoogLeNet, and VGG-16) on the smartphone. Significant reductions in model size, runtime, and energy consumption are obtained, at the cost of small loss in accuracy. In addition, we address the important implementation level issue on 1?1 convolution, which is a key operation of inception module of GoogLeNet as well as CNNs compressed by our proposed scheme.

arXiv:1602.02505공식 코드 · Papers with Code

Binarized Neural Networks

We introduce a method to train Binarized Neural Networks (BNNs) - neural networks with binary weights and activations at run-time and when computing the parameters' gradient at train-time. We conduct two sets of experiments, each based on a different framework, namely Torch7 and Theano, where we train BNNs on MNIST, CIFAR-10 and SVHN, and achieve nearly state-of-the-art results. During the forward pass, BNNs drastically reduce memory size and accesses, and replace most arithmetic operations with bit-wise operations, which might lead to a great increase in power-efficiency. Last but not least, we wrote a binary matrix multiplication GPU kernel with which it is possible to run our MNIST BNN 7 times faster than with an unoptimized GPU kernel, without suffering any loss in classification accuracy. The code for training and running our BNNs is available.

arXiv:1511.08228공식 코드 · Papers with Code

Neural GPUs Learn Algorithms

Learning an algorithm from examples is a fundamental problem that has been widely studied. Recently it has been addressed using neural networks, in particular by Neural Turing Machines (NTMs). These are fully differentiable computers that use backpropagation to learn their own programming. Despite their appeal NTMs have a weakness that is caused by their sequential nature: they are not parallel and are are hard to train due to their large depth when unfolded. We present a neural network architecture to address this problem: the Neural GPU. It is based on a type of convolutional gated recurrent unit and, like the NTM, is computationally universal. Unlike the NTM, the Neural GPU is highly parallel which makes it easier to train and efficient to run. An essential property of algorithms is their ability to handle inputs of arbitrary size. We show that the Neural GPU can be trained on short instances of an algorithmic task and successfully generalize to long instances. We verified it on a number of tasks including long addition and long multiplication of numbers represented in binary. We train the Neural GPU on numbers with upto 20 bits and observe no errors whatsoever while testing it, even on much longer numbers. To achieve these results we introduce a technique for training deep recurrent networks: parameter sharing relaxation. We also found a small amount of dropout and gradient noise to have a large positive effect on learning and generalization.