PyTorchKR Papers

최신 논문

총 5,575편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:1711.06491공식 코드 · Papers with Code

High-Resolution Deep Convolutional Generative Adversarial Networks

Generative Adversarial Networks (GANs) convergence in a high-resolution setting with a computational constrain of GPU memory capacity (from 12GB to 24 GB) has been beset with difficulty due to the known lack of convergence rate stability. In order to boost network convergence of DCGAN (Deep Convolutional Generative Adversarial Networks) and achieve good-looking high-resolution results we propose a new layered network structure, HDCGAN, that incorporates current state-of-the-art techniques for this effect. A novel dataset, Curt\'o & Zarza, containing human faces from different ethnical groups in a wide variety of illumination conditions and image resolutions is introduced. Curt\'o is enhanced with HDCGAN synthetic images, thus being the first GAN augmented face dataset. We conduct extensive experiments on CelebA (MS-SSIM 0.1978 and Distance of Fr\'echet 8.77) and Curt\'o.

arXiv:1805.08687

Attaining human-level performance with atlas location autocontext for anatomical landmark detection in 3D CT data

We present an efficient neural network method for locating anatomical landmarks in 3D medical CT scans, using atlas location autocontext in order to learn long-range spatial context. Location predictions are made by regression to Gaussian heatmaps, one heatmap per landmark. This system allows patchwise application of a shallow network, thus enabling multiple volumetric heatmaps to be predicted concurrently without prohibitive GPU memory requirements. Further, the system allows inter-landmark spatial relationships to be exploited using a simple overdetermined affine mapping that is robust to detection failures and occlusion or partial views. Evaluation is performed for 22 landmarks defined on a range of structures in head CT scans. Models are trained and validated on 201 scans. Over the final test set of 20 scans which was independently annotated by 2 human annotators, the neural network reaches an accuracy which matches the annotator variability, with similar human and machine patterns of variability across landmark classes.

arXiv:1805.06386

Neural Multi-scale Image Compression

This study presents a new lossy image compression method that utilizes the multi-scale features of natural images. Our model consists of two networks: multi-scale lossy autoencoder and parallel multi-scale lossless coder. The multi-scale lossy autoencoder extracts the multi-scale image features to quantized variables and the parallel multi-scale lossless coder enables rapid and accurate lossless coding of the quantized variables via encoding/decoding the variables in parallel. Our proposed model achieves comparable performance to the state-of-the-art model on Kodak and RAISE-1k dataset images, and it encodes a PNG image of size $768 \times 512$ in 70 ms with a single GPU and a single CPU process and decodes it into a high-fidelity image in approximately 200 ms.

arXiv:1805.06383

Composing Finite State Transducers on GPUs

Weighted finite-state transducers (FSTs) are frequently used in language processing to handle tasks such as part-of-speech tagging and speech recognition. There has been previous work using multiple CPU cores to accelerate finite state algorithms, but limited attention has been given to parallel graphics processing unit (GPU) implementations. In this paper, we introduce the first (to our knowledge) GPU implementation of the FST composition operation, and we also discuss the optimizations used to achieve the best performance on this architecture. We show that our approach obtains speedups of up to 6x over our serial implementation and 4.5x over OpenFST.

arXiv:1803.01364

SAFE: Spectral Evolution Analysis Feature Extraction for Non-Stationary Time Series Prediction

This paper presents a practical approach for detecting non-stationarity in time series prediction. This method is called SAFE and works by monitoring the evolution of the spectral contents of time series through a distance function. This method is designed to work in combination with state-of-the-art machine learning methods in real time by informing the online predictors to perform necessary adaptation when a non-stationarity presents. We also propose an algorithm to proportionally include some past data in the adaption process to overcome the Catastrophic Forgetting problem. To validate our hypothesis and test the effectiveness of our approach, we present comprehensive experiments in different elements of the approach involving artificial and real-world datasets. The experiments show that the proposed method is able to significantly save computational resources in term of processor or GPU cycles while maintaining high prediction performances.

arXiv:1712.00960공식 코드 · Papers with Code

FSSD: Feature Fusion Single Shot Multibox Detector

SSD (Single Shot Multibox Detector) is one of the best object detection algorithms with both high accuracy and fast speed. However, SSD's feature pyramid detection method makes it hard to fuse the features from different scales. In this paper, we proposed FSSD (Feature Fusion Single Shot Multibox Detector), an enhanced SSD with a novel and lightweight feature fusion module which can improve the performance significantly over SSD with just a little speed drop. In the feature fusion module, features from different layers with different scales are concatenated together, followed by some down-sampling blocks to generate new feature pyramid, which will be fed to multibox detectors to predict the final detection results. On the Pascal VOC 2007 test, our network can achieve 82.7 mAP (mean average precision) at the speed of 65.8 FPS (frame per second) with the input size 300$\times$300 using a single Nvidia 1080Ti GPU. In addition, our result on COCO is also better than the conventional SSD with a large margin. Our FSSD outperforms a lot of state-of-the-art object detection algorithms in both aspects of accuracy and speed. Code is available at https://github.com/lzx1413/CAFFE_SSD/tree/fssd.

arXiv:1805.07440

Neural Architecture Search using Deep Neural Networks and Monte Carlo Tree Search

Neural Architecture Search (NAS) has shown great success in automating the design of neural networks, but the prohibitive amount of computations behind current NAS methods requires further investigations in improving the sample efficiency and the network evaluation cost to get better results in a shorter time. In this paper, we present a novel scalable Monte Carlo Tree Search (MCTS) based NAS agent, named AlphaX, to tackle these two aspects. AlphaX improves the search efficiency by adaptively balancing the exploration and exploitation at the state level, and by a Meta-Deep Neural Network (DNN) to predict network accuracies for biasing the search toward a promising region. To amortize the network evaluation cost, AlphaX accelerates MCTS rollouts with a distributed design and reduces the number of epochs in evaluating a network by transfer learning, which is guided with the tree structure in MCTS. In 12 GPU days and 1000 samples, AlphaX found an architecture that reaches 97.84\% top-1 accuracy on CIFAR-10, and 75.5\% top-1 accuracy on ImageNet, exceeding SOTA NAS methods in both the accuracy and sampling efficiency. Particularly, we also evaluate AlphaX on NASBench-101, a large scale NAS dataset; AlphaX is 3x and 2.8x more sample efficient than Random Search and Regularized Evolution in finding the global optimum. Finally, we show the searched architecture improves a variety of vision applications from Neural Style Transfer, to Image Captioning and Object Detection.

arXiv:1802.02326

A Novel Co-design Peta-scale Heterogeneous Cluster for Deep Learning Training

Large scale deep Convolution Neural Networks (CNNs) increasingly demands the computing power. It is key for researchers to own a great powerful computing platform to leverage deep learning (DL) advancing.On the other hand, as the commonly-used accelerator, the commodity GPUs cards of new generations are more and more expensive. Consequently, it is of importance to design an affordable distributed heterogeneous system that provides powerful computational capacity and develop a well-suited software that efficiently utilizes its computational capacity. In this paper, we present our co-design distributed system including a peta-scale GPU cluster, called "Manoa". Based on properties and topology of Manoa, we first propose job server framework and implement it, named "MiMatrix". The central node of MiMatrix, referred to as the job server, undertakes all of controlling, scheduling and monitoring, and I/O tasks without weight data transfer for AllReduce processing in each iteration. Therefore, MiMatrix intrinsically solves the bandwidth bottleneck of central node in parameter server framework that is widely used in distributed DL tasks. Meanwhile, we also propose a new AllReduce algorithm, GPUDirect RDMA-Aware AllReduce~(GDRAA), in which both computation and handshake message are O(1) and the number of synchronization is two in each iteration that is a theoretical minimum number. Owe to the dedicated co-design distributed system, MiMatrix efficiently makes use of the Manoa's computational capacity and bandwidth. We benchmark Manoa Resnet50 and Resenet101 on Imagenet-1K dataset. Some of results have demonstrated state-of-the-art.

arXiv:1802.04799

TVM: An Automated End-to-End Optimizing Compiler for Deep Learning

There is an increasing need to bring machine learning to a wide diversity of hardware devices. Current frameworks rely on vendor-specific operator libraries and optimize for a narrow range of server-class GPUs. Deploying workloads to new platforms -- such as mobile phones, embedded devices, and accelerators (e.g., FPGAs, ASICs) -- requires significant manual effort. We propose TVM, a compiler that exposes graph-level and operator-level optimizations to provide performance portability to deep learning workloads across diverse hardware back-ends. TVM solves optimization challenges specific to deep learning, such as high-level operator fusion, mapping to arbitrary hardware primitives, and memory latency hiding. It also automates optimization of low-level programs to hardware characteristics by employing a novel, learning-based cost modeling method for rapid exploration of code optimizations. Experimental results show that TVM delivers performance across hardware back-ends that are competitive with state-of-the-art, hand-tuned libraries for low-power CPU, mobile GPU, and server-class GPUs. We also demonstrate TVM's ability to target new accelerator back-ends, such as the FPGA-based generic deep learning accelerator. The system is open sourced and in production use inside several major companies.

arXiv:1805.08166

Learning to Optimize Tensor Programs

We introduce a learning-based framework to optimize tensor programs for deep learning workloads. Efficient implementations of tensor operators, such as matrix multiplication and high dimensional convolution, are key enablers of effective deep learning systems. However, existing systems rely on manually optimized libraries such as cuDNN where only a narrow range of server class GPUs are well-supported. The reliance on hardware-specific operator libraries limits the applicability of high-level graph optimizations and incurs significant engineering costs when deploying to new hardware targets. We use learning to remove this engineering burden. We learn domain-specific statistical cost models to guide the search of tensor operator implementations over billions of possible program variants. We further accelerate the search by effective model transfer across workloads. Experimental results show that our framework delivers performance competitive with state-of-the-art hand-tuned libraries for low-power CPU, mobile GPU, and server-class GPU.

arXiv:1805.08009

Object Detection in Equirectangular Panorama

We introduced a high-resolution equirectangular panorama (360-degree, virtual reality) dataset for object detection and propose a multi-projection variant of YOLO detector. The main challenge with equirectangular panorama image are i) the lack of annotated training data, ii) high-resolution imagery and iii) severe geometric distortions of objects near the panorama projection poles. In this work, we solve the challenges by i) using training examples available in the "conventional datasets" (ImageNet and COCO), ii) employing only low-resolution images that require only moderate GPU computing power and memory, and iii) our multi-projection YOLO handles projection distortions by making multiple stereographic sub-projections. In our experiments, YOLO outperforms the other state-of-art detector, Faster RCNN and our multi-projection YOLO achieves the best accuracy with low-resolution input.

arXiv:1805.07978

Energy-Efficient Inference Accelerator for Memory-Augmented Neural Networks on an FPGA

Memory-augmented neural networks (MANNs) are designed for question-answering tasks. It is difficult to run a MANN effectively on accelerators designed for other neural networks (NNs), in particular on mobile devices, because MANNs require recurrent data paths and various types of operations related to external memory access. We implement an accelerator for MANNs on a field-programmable gate array (FPGA) based on a data flow architecture. Inference times are also reduced by inference thresholding, which is a data-based maximum inner-product search specialized for natural language tasks. Measurements on the bAbI data show that the energy efficiency of the accelerator (FLOPS/kJ) was higher than that of an NVIDIA TITAN V GPU by a factor of about 125, increasing to 140 with inference thresholding

arXiv:1805.08308공식 코드 · Papers with Code

geomstats: a Python Package for Riemannian Geometry in Machine Learning

We introduce geomstats, a python package that performs computations on manifolds such as hyperspheres, hyperbolic spaces, spaces of symmetric positive definite matrices and Lie groups of transformations. We provide efficient and extensively unit-tested implementations of these manifolds, together with useful Riemannian metrics and associated Exponential and Logarithm maps. The corresponding geodesic distances provide a range of intuitive choices of Machine Learning loss functions. We also give the corresponding Riemannian gradients. The operations implemented in geomstats are available with different computing backends such as numpy, tensorflow and keras. We have enabled GPU implementation and integrated geomstats manifold computations into keras deep learning framework. This paper also presents a review of manifolds in machine learning and an overview of the geomstats package with examples demonstrating its use for efficient and user-friendly Riemannian geometry.

arXiv:1805.08899

Echo: Compiler-based GPU Memory Footprint Reduction for LSTM RNN Training

The Long-Short-Term-Memory Recurrent Neural Networks (LSTM RNNs) are a popular class of machine learning models for analyzing sequential data. Their training on modern GPUs, however, is limited by the GPU memory capacity. Our profiling results of the LSTM RNN-based Neural Machine Translation (NMT) model reveal that feature maps of the attention and RNN layers form the memory bottleneck and runtime is unevenly distributed across different layers when training on GPUs. Based on these two observations, we propose to recompute the feature maps rather than stashing them persistently in the GPU memory. While the idea of feature map recomputation has been considered before, existing solutions fail to deliver satisfactory footprint reduction, as they do not address two key challenges. For each feature map recomputation to be effective and efficient, its effect on (1) the total memory footprint, and (2) the total execution time has to be carefully estimated. To this end, we propose *Echo*, a new compiler-based optimization scheme that addresses the first challenge with a practical mechanism that estimates the memory benefits of recomputation over the entire computation graph, and the second challenge by non-conservatively estimating the recomputation overhead leveraging layer specifics. *Echo* reduces the GPU memory footprint automatically and transparently without any changes required to the training source code, and is effective for models beyond LSTM RNNs. We evaluate *Echo* on numerous state-of-the-art machine learning workloads on real systems with modern GPUs and observe footprint reduction ratios of 1.89X on average and 3.13X maximum. Such reduction can be converted into faster training with a larger batch size, savings in GPU energy consumption (e.g., training with one GPU as fast as with four), and/or an increase in the maximum number of layers under the same GPU memory budget.

arXiv:1803.06744

Fast Neural Architecture Construction using EnvelopeNets

Fast Neural Architecture Construction (NAC) is a method to construct deep network architectures by pruning and expansion of a base network. In recent years, several automated search methods for neural network architectures have been proposed using methods such as evolutionary algorithms and reinforcement learning. These methods use a single scalar objective function (usually accuracy) that is evaluated after a full training and evaluation cycle. In contrast NAC directly compares the utility of different filters using statistics derived from filter featuremaps reach a state where the utility of different filters within a network can be compared and hence can be used to construct networks. The training epochs needed for filters within a network to reach this state is much less than the training epochs needed for the accuracy of a network to stabilize. NAC exploits this finding to construct convolutional neural nets (CNNs) with close to state of the art accuracy, in < 1 GPU day, faster than most of the current neural architecture search methods. The constructed networks show close to state of the art performance on the image classification problem on well known datasets (CIFAR-10, ImageNet) and consistently show better performance than hand constructed and randomly generated networks of the same depth, operators and approximately the same number of parameters.

arXiv:1711.10314공식 코드 · Papers with Code

Crossmodal Attentive Skill Learner

This paper presents the Crossmodal Attentive Skill Learner (CASL), integrated with the recently-introduced Asynchronous Advantage Option-Critic (A2OC) architecture [Harb et al., 2017] to enable hierarchical reinforcement learning across multiple sensory inputs. We provide concrete examples where the approach not only improves performance in a single task, but accelerates transfer to new tasks. We demonstrate the attention mechanism anticipates and identifies useful latent features, while filtering irrelevant sensor modalities during execution. We modify the Arcade Learning Environment [Bellemare et al., 2013] to support audio queries, and conduct evaluations of crossmodal learning in the Atari 2600 game Amidar. Finally, building on the recent work of Babaeizadeh et al. [2017], we open-source a fast hybrid CPU-GPU implementation of CASL.

arXiv:1805.08995

GPU Accelerated Cascade Hashing Image Matching for Large Scale 3D Reconstruction

Image feature point matching is a key step in Structure from Motion(SFM). However, it is becoming more and more time consuming because the number of images is getting larger and larger. In this paper, we proposed a GPU accelerated image matching method with improved Cascade Hashing. Firstly, we propose a Disk-Memory-GPU data exchange strategy and optimize the load order of data, so that the proposed method can deal with big data. Next, we parallelize the Cascade Hashing method on GPU. An improved parallel reduction and an improved parallel hashing ranking are proposed to fulfill this task. Finally, extensive experiments show that our image matching is about 20 times faster than SiftGPU on the same graphics card, nearly 100 times faster than the CPU CasHash method and hundreds of times faster than the CPU Kd-Tree based matching method. Further more, we introduce the epipolar constraint to the proposed method, and use the epipolar geometry to guide the feature matching procedure, which further reduces the matching cost.

arXiv:1805.10174

f-CNN$^{\text{x}}$: A Toolflow for Mapping Multi-CNN Applications on FPGAs

The predictive power of Convolutional Neural Networks (CNNs) has been an integral factor for emerging latency-sensitive applications, such as autonomous drones and vehicles. Such systems employ multiple CNNs, each one trained for a particular task. The efficient mapping of multiple CNNs on a single FPGA device is a challenging task as the allocation of compute resources and external memory bandwidth needs to be optimised at design time. This paper proposes f-CNN$^{\text{x}}$, an automated toolflow for the optimised mapping of multiple CNNs on FPGAs, comprising a novel multi-CNN hardware architecture together with an automated design space exploration method that considers the user-specified performance requirements for each model to allocate compute resources and generate a synthesisable accelerator. Moreover, f-CNN$^{\text{x}}$ employs a novel scheduling algorithm that alleviates the limitations of the memory bandwidth contention between CNNs and sustains the high utilisation of the architecture. Experimental evaluation shows that f-CNN$^{\text{x}}$'s designs outperform contention-unaware FPGA mappings by up to 50% and deliver up to 6.8x higher performance-per-Watt over highly optimised GPU designs for multi-CNN systems.

arXiv:1805.10958

Discrete flow posteriors for variational inference in discrete dynamical systems

Each training step for a variational autoencoder (VAE) requires us to sample from the approximate posterior, so we usually choose simple (e.g. factorised) approximate posteriors in which sampling is an efficient computation that fully exploits GPU parallelism. However, such simple approximate posteriors are often insufficient, as they eliminate statistical dependencies in the posterior. While it is possible to use normalizing flow approximate posteriors for continuous latents, some problems have discrete latents and strong statistical dependencies. The most natural approach to model these dependencies is an autoregressive distribution, but sampling from such distributions is inherently sequential and thus slow. We develop a fast, parallel sampling procedure for autoregressive distributions based on fixed-point iterations which enables efficient and accurate variational inference in discrete state-space latent variable dynamical systems. To optimize the variational bound, we considered two ways to evaluate probabilities: inserting the relaxed samples directly into the pmf for the discrete distribution, or converting to continuous logistic latent variables and interpreting the K-step fixed-point iterations as a normalizing flow. We found that converting to continuous latent variables gave considerable additional scope for mismatch between the true and approximate posteriors, which resulted in biased inferences, we thus used the former approach. Using our fast sampling procedure, we were able to realize the benefits of correlated posteriors, including accurate uncertainty estimates for one cell, and accurate connectivity estimates for multiple cells, in an order of magnitude less time.

arXiv:1805.10737

Deep Adversarial Context-Aware Landmark Detection for Ultrasound Imaging

Real-time localization of prostate gland in trans-rectal ultrasound images is a key technology that is required to automate the ultrasound guided prostate biopsy procedures. In this paper, we propose a new deep learning based approach which is aimed at localizing several prostate landmarks efficiently and robustly. We propose a multitask learning approach primarily to make the overall algorithm more contextually aware. In this approach, we not only consider the explicit learning of landmark locations, but also build-in a mechanism to learn the contour of the prostate. This multitask learning is further coupled with an adversarial arm to promote the generation of feasible structures. We have trained this network using ~4000 labeled trans-rectal ultrasound images and tested on an independent set of images with ground truth landmark locations. We have achieved an overall Dice score of 92.6% for the adversarially trained multitask approach, which is significantly better than the Dice score of 88.3% obtained by only learning of landmark locations. The overall mean distance error using the adversarial multitask approach has also improved by 20% while reducing the standard deviation of the error compared to learning landmark locations only. In terms of computational complexity both approaches can process the images in real-time using standard computer with a standard CUDA enabled GPU.