Modern Systems on Chip (SoC), almost as a rule, require accelerators for achieving energy efficiency and high performance for specific tasks that are not necessarily well suited for execution in standard processing units. Considering the broad range of applications and necessity for specialization, the design of SoCs has thus become expressively more challenging. In this paper, we put forward the concept of G-GPU, a general-purpose GPU-like accelerator that is not application-specific but still gives benefits in energy efficiency and throughput. Furthermore, we have identified an existing gap for these accelerators in ASIC, for which no known automated generation platform/tool exists. Our solution, called GPUPlanner, is an open-source generator of accelerators, from RTL to GDSII, that addresses this gap. Our analysis results show that our automatically generated G-GPU designs are remarkably efficient when compared against the popular CPU architecture RISC-V, presenting speed-ups of up to 223 times in raw performance and up to 11 times when the metric is performance derated by area. These results are achieved by executing a design space exploration of the GPU-like accelerators, where the memory hierarchy is broken in a smart fashion and the logic is pipelined on demand. Finally, tapeout-ready layouts of the G-GPU in 65nm CMOS are presented.
相關內容
Hamilton and Moitra (2021) showed that, in certain regimes, it is not possible to accelerate Riemannian gradient descent in the hyperbolic plane if we restrict ourselves to algorithms which make queries in a (large) bounded domain and which receive gradients and function values corrupted by a (small) amount of noise. We show that acceleration remains unachievable for any deterministic algorithm which receives exact gradient and function-value information (unbounded queries, no noise). Our results hold for the classes of strongly and nonstrongly geodesically convex functions, and for a large class of Hadamard manifolds including hyperbolic spaces and the symmetric space $\mathrm{SL}(n) / \mathrm{SO}(n)$ of positive definite $n \times n$ matrices of determinant one. This cements a surprising gap between the complexity of convex optimization and geodesically convex optimization: for hyperbolic spaces, Riemannian gradient descent is optimal on the class of smooth and and strongly geodesically convex functions, in the regime where the condition number scales with the radius of the optimization domain. The key idea for proving the lower bound consists of perturbing the hard functions of Hamilton and Moitra (2021) with sums of bump functions chosen by a resisting oracle.
Sparse-Matrix Dense-Matrix multiplication (SpMM) is the key operator for a wide range of applications, including scientific computing, graph processing, and deep learning. Architecting accelerators for SpMM is faced with three challenges - (1) the random memory accessing and unbalanced load in processing because of random distribution of elements in sparse matrices, (2) inefficient data handling of the large matrices which can not be fit on-chip, and (3) anon-general-purpose accelerator design where one accelerator can only process a fixed-size problem. In this paper, we present Sextans, an accelerator for general-purpose SpMM processing. Sextans accelerator features (1) fast random access using on-chip memory, (2) streaming access to off-chip large matrices, (3) PE-aware non-zero scheduling for balanced workload with an II=1 pipeline, and (4) hardware flexibility to enable prototyping the hardware once to support SpMMs of different size as a general-purpose accelerator. We leverage high bandwidth memory (HBM) for the efficient accessing of both sparse and dense matrices. In the evaluation, we present an FPGA prototype Sextans which is executable on a Xilinx U280 HBM FPGA board and a projected prototype Sextans-P with higher bandwidth comparable to V100 and more frequency optimization. We conduct a comprehensive evaluation on 1,400 SpMMs on a wide range of sparse matrices including 50 matrices from SNAP and 150 from SuiteSparse. WecompareSextanswith NVIDIA K80 and V100 GPUs.Sextansachieves a 2.50x geomean speedup over K80 GPU andSextans-Pachieves a 1.14x geomean speedup over V100 GPU (4.94x over K80). The code is available at //github.com/linghaosong/Sextans.
Rule-based classifier, that extract a subset of induced rules to efficiently learn/mine while preserving the discernibility information, plays a crucial role in human-explainable artificial intelligence. However, in this era of big data, rule induction on the whole datasets is computationally intensive. So far, to the best of our knowledge, no known method focusing on accelerating rule induction has been reported. This is first study to consider the acceleration technique to reduce the scale of computation in rule induction. We propose an accelerator for rule induction based on fuzzy rough theory; the accelerator can avoid redundant computation and accelerate the building of a rule classifier. First, a rule induction method based on consistence degree, called Consistence-based Value Reduction (CVR), is proposed and used as basis to accelerate. Second, we introduce a compacted search space termed Key Set, which only contains the key instances required to update the induced rule, to conduct value reduction. The monotonicity of Key Set ensures the feasibility of our accelerator. Third, a rule-induction accelerator is designed based on Key Set, and it is theoretically guaranteed to display the same results as the unaccelerated version. Specifically, the rank preservation property of Key Set ensures consistency between the rule induction achieved by the accelerator and the unaccelerated method. Finally, extensive experiments demonstrate that the proposed accelerator can perform remarkably faster than the unaccelerated rule-based classifier methods, especially on datasets with numerous instances.
Over the past several years, new machine learning accelerators were being announced and released every month for a variety of applications from speech recognition, video object detection, assisted driving, and many data center applications. This paper updates the survey of AI accelerators and processors from past two years. This paper collects and summarizes the current commercial accelerators that have been publicly announced with peak performance and power consumption numbers. The performance and power values are plotted on a scatter graph, and a number of dimensions and observations from the trends on this plot are again discussed and analyzed. This year, we also compile a list of benchmarking performance results and compute the computational efficiency with respect to peak performance.
Transformers have been successful for many natural language processing tasks. However, applying transformers to the video domain for tasks such as long-term video generation and scene understanding has remained elusive due to the high computational complexity and the lack of natural tokenization. In this paper, we propose the Object-Centric Video Transformer (OCVT) which utilizes an object-centric approach for decomposing scenes into tokens suitable for use in a generative video transformer. By factoring the video into objects, our fully unsupervised model is able to learn complex spatio-temporal dynamics of multiple interacting objects in a scene and generate future frames of the video. Our model is also significantly more memory-efficient than pixel-based models and thus able to train on videos of length up to 70 frames with a single 48GB GPU. We compare our model with previous RNN-based approaches as well as other possible video transformer baselines. We demonstrate OCVT performs well when compared to baselines in generating future frames. OCVT also develops useful representations for video reasoning, achieving start-of-the-art performance on the CATER task.
Most Deep Reinforcement Learning (Deep RL) algorithms require a prohibitively large number of training samples for learning complex tasks. Many recent works on speeding up Deep RL have focused on distributed training and simulation. While distributed training is often done on the GPU, simulation is not. In this work, we propose using GPU-accelerated RL simulations as an alternative to CPU ones. Using NVIDIA Flex, a GPU-based physics engine, we show promising speed-ups of learning various continuous-control, locomotion tasks. With one GPU and CPU core, we are able to train the Humanoid running task in less than 20 minutes, using 10-1000x fewer CPU cores than previous works. We also demonstrate the scalability of our simulator to multi-GPU settings to train more challenging locomotion tasks.
Fully convolutional neural network (FCN) has been dominating the game of face detection task for a few years with its congenital capability of sliding-window-searching with shared kernels, which boiled down all the redundant calculation, and most recent state-of-the-art methods such as Faster-RCNN, SSD, YOLO and FPN use FCN as their backbone. So here comes one question: Can we find a universal strategy to further accelerate FCN with higher accuracy, so could accelerate all the recent FCN-based methods? To analyze this, we decompose the face searching space into two orthogonal directions, `scale' and `spatial'. Only a few coordinates in the space expanded by the two base vectors indicate foreground. So if FCN could ignore most of the other points, the searching space and false alarm should be significantly boiled down. Based on this philosophy, a novel method named scale estimation and spatial attention proposal ($S^2AP$) is proposed to pay attention to some specific scales and valid locations in the image pyramid. Furthermore, we adopt a masked-convolution operation based on the attention result to accelerate FCN calculation. Experiments show that FCN-based method RPN can be accelerated by about $4\times$ with the help of $S^2AP$ and masked-FCN and at the same time it can also achieve the state-of-the-art on FDDB, AFW and MALF face detection benchmarks as well.
When a recurrent neural network language model is used for caption generation, the image information can be fed to the neural network either by directly incorporating it in the RNN -- conditioning the language model by `injecting' image features -- or in a layer following the RNN -- conditioning the language model by `merging' image features. While both options are attested in the literature, there is as yet no systematic comparison between the two. In this paper we empirically show that it is not especially detrimental to performance whether one architecture is used or another. The merge architecture does have practical advantages, as conditioning by merging allows the RNN's hidden state vector to shrink in size by up to four times. Our results suggest that the visual and linguistic modalities for caption generation need not be jointly encoded by the RNN as that yields large, memory-intensive models with few tangible advantages in performance; rather, the multimodal integration should be delayed to a subsequent stage.
Latent Dirichlet Allocation(LDA) is a popular topic model. Given the fact that the input corpus of LDA algorithms consists of millions to billions of tokens, the LDA training process is very time-consuming, which may prevent the usage of LDA in many scenarios, e.g., online service. GPUs have benefited modern machine learning algorithms and big data analysis as they can provide high memory bandwidth and computation power. Therefore, many frameworks, e.g. Ten- sorFlow, Caffe, CNTK, support to use GPUs for accelerating the popular machine learning data-intensive algorithms. However, we observe that LDA solutions on GPUs are not satisfying. In this paper, we present CuLDA_CGS, a GPU-based efficient and scalable approach to accelerate large-scale LDA problems. CuLDA_CGS is designed to efficiently solve LDA problems at high throughput. To it, we first delicately design workload partition and synchronization mechanism to exploit the benefits of mul- tiple GPUs. Then, we offload the LDA sampling process to each individual GPU by optimizing from the sampling algorithm, par- allelization, and data compression perspectives. Evaluations show that compared with state-of-the-art LDA solutions, CuLDA_CGS outperforms them by a large margin (up to 7.3X) on a single GPU. CuLDA_CGS is able to achieve extra 3.0X speedup on 4 GPUs. The source code is publicly available on //github.com/cuMF/ CuLDA_CGS.
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 (CZ), containing human faces from different ethnical groups in a wide variety of illumination conditions and image resolutions is introduced. We conduct extensive experiments on CelebA and CZ.
小貼士
登錄享
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191