PyTorchKR Papers

최신 논문

총 93편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2201.08475공식 코드 · Papers with Code

GenGNN: A Generic FPGA Framework for Graph Neural Network Acceleration

Graph neural networks (GNNs) have recently exploded in popularity thanks to their broad applicability to ubiquitous graph-related problems such as quantum chemistry, drug discovery, and high energy physics. However, meeting demand for novel GNN models and fast inference simultaneously is challenging because of the gap between the difficulty in developing efficient FPGA accelerators and the rapid pace of creation of new GNN models. Prior art focuses on the acceleration of specific classes of GNNs but lacks the generality to work across existing models or to extend to new and emerging GNN models. In this work, we propose a generic GNN acceleration framework using High-Level Synthesis (HLS), named GenGNN, with two-fold goals. First, we aim to deliver ultra-fast GNN inference without any graph pre-processing for real-time requirements. Second, we aim to support a diverse set of GNN models with the extensibility to flexibly adapt to new models. The framework features an optimized message-passing structure applicable to all models, combined with a rich library of model-specific components. We verify our implementation on-board on the Xilinx Alveo U50 FPGA and observe a speed-up of up to 25x against CPU (6226R) baseline and 13x against GPU (A6000) baseline. Our HLS code will be open-source on GitHub upon acceptance.

arXiv:2201.06878공식 코드 · Papers with Code

Hardware-Efficient Deconvolution-Based GAN for Edge Computing

Generative Adversarial Networks (GAN) are cutting-edge algorithms for generating new data samples based on the learned data distribution. However, its performance comes at a significant cost in terms of computation and memory requirements. In this paper, we proposed an HW/SW co-design approach for training quantized deconvolution GAN (QDCGAN) implemented on FPGA using a scalable streaming dataflow architecture capable of achieving higher throughput versus resource utilization trade-off. The developed accelerator is based on an efficient deconvolution engine that offers high parallelism with respect to scaling factors for GAN-based edge computing. Furthermore, various precisions, datasets, and network scalability were analyzed for low-power inference on resource-constrained platforms. Lastly, an end-to-end open-source framework is provided for training, implementation, state-space exploration, and scaling the inference using Vivado high-level synthesis for Xilinx SoC-FPGAs, and a comparison testbed with Jetson Nano.

arXiv:2201.06848

High-Level Synthesis Performance Prediction using GNNs: Benchmarking, Modeling, and Advancing

Agile hardware development requires fast and accurate circuit quality evaluation from early design stages. Existing work of high-level synthesis (HLS) performance prediction usually needs extensive feature engineering after the synthesis process. To expedite circuit evaluation from as earlier design stage as possible, we propose a rapid and accurate performance modeling, exploiting the representation power of graph neural networks (GNNs) by representing C/C++ programs as graphs. The contribution of this work is three-fold. First, we build a standard benchmark containing 40k C synthesizable programs, which includes both synthetic programs and three sets of real-world HLS benchmarks. Each program is implemented on FPGA to generate ground-truth performance metrics. Second, we formally formulate the HLS performance prediction problem on graphs, and propose multiple modeling strategies with GNNs that leverage different trade-offs between prediction timeliness (early/late prediction) and accuracy. Third, we further propose a novel hierarchical GNN that does not sacrifice timeliness but largely improves prediction accuracy, significantly outperforming HLS tools. We apply extensive evaluations for both synthetic and unseen real-case programs; our proposed predictor largely outperforms HLS by up to 40X and excels existing predictors by 2X to 5X in terms of resource usage and timing prediction.

arXiv:2201.06618

VAQF: Fully Automatic Software-Hardware Co-Design Framework for Low-Bit Vision Transformer

The transformer architectures with attention mechanisms have obtained success in Nature Language Processing (NLP), and Vision Transformers (ViTs) have recently extended the application domains to various vision tasks. While achieving high performance, ViTs suffer from large model size and high computation complexity that hinders the deployment of them on edge devices. To achieve high throughput on hardware and preserve the model accuracy simultaneously, we propose VAQF, a framework that builds inference accelerators on FPGA platforms for quantized ViTs with binary weights and low-precision activations. Given the model structure and the desired frame rate, VAQF will automatically output the required quantization precision for activations as well as the optimized parameter settings of the accelerator that fulfill the hardware requirements. The implementations are developed with Vivado High-Level Synthesis (HLS) on the Xilinx ZCU102 FPGA board, and the evaluation results with the DeiT-base model indicate that a frame rate requirement of 24 frames per second (FPS) is satisfied with 8-bit activation quantization, and a target of 30 FPS is met with 6-bit activation quantization. To the best of our knowledge, this is the first time quantization has been incorporated into ViT acceleration on FPGAs with the help of a fully automatic framework to guide the quantization strategy on the software side and the accelerator implementations on the hardware side given the target frame rate. Very small compilation time cost is incurred compared with quantization training, and the generated accelerators show the capability of achieving real-time execution for state-of-the-art ViT models on FPGAs.

arXiv:2111.14767

A Graph Deep Learning Framework for High-Level Synthesis Design Space Exploration

The design of efficient hardware accelerators for high-throughput data-processing applications, e.g., deep neural networks, is a challenging task in computer architecture design. In this regard, High-Level Synthesis (HLS) emerges as a solution for fast prototyping application-specific hardware starting from a behavioural description of the application computational flow. This Design-Space Exploration (DSE) aims at identifying Pareto optimal synthesis configurations whose exhaustive search is often unfeasible due to the design-space dimensionality and the prohibitive computational cost of the synthesis process. Within this framework, we effectively and efficiently address the design problem by proposing, for the first time in the literature, graph neural networks that jointly predict acceleration performance and hardware costs of a synthesized behavioral specification given optimization directives. The learned model can be used to rapidly approach the Pareto curve by guiding the DSE, taking into account performance and cost estimates. The proposed method outperforms traditional HLS-driven DSE approaches, by accounting for arbitrary length of computer programs and the invariant properties of the input. We propose a novel hybrid control and data flow graph representation that enables training the graph neural network on specifications of different hardware accelerators; the methodology naturally transfers to unseen data-processing applications too. Moreover, we show that our approach achieves prediction accuracy comparable with that of commonly used simulators without having access to analytical models of the HLS compiler and the target FPGA, while being orders of magnitude faster. Finally, the learned representation can be exploited for DSE in unexplored configuration spaces by fine-tuning on a small number of samples from the new target domain.

arXiv:2111.08848

Enabling Automated FPGA Accelerator Optimization Using Graph Neural Networks

High-level synthesis (HLS) has freed the computer architects from developing their designs in a very low-level language and needing to exactly specify how the data should be transferred in register-level. With the help of HLS, the hardware designers must describe only a high-level behavioral flow of the design. Despite this, it still can take weeks to develop a high-performance architecture mainly because there are many design choices at a higher level that requires more time to explore. It also takes several minutes to hours to get feedback from the HLS tool on the quality of each design candidate. In this paper, we propose to solve this problem by modeling the HLS tool with a graph neural network (GNN) that is trained to be used for a wide range of applications. The experimental results demonstrate that by employing the GNN-based model, we are able to estimate the quality of design in milliseconds with high accuracy which can help us search through the solution space very quickly.

arXiv:2109.14840

A system on chip for melanoma detection using FPGA-based SVM classifier

Support Vector Machine (SVM) is a robust machine learning model that shows high accuracy with different classification problems, and is widely used for various embedded applications. However , implementation of embedded SVM classifiers is challenging, due to the inherent complicated computations required. This motivates implementing the SVM on hardware platforms for achieving high performance computing at low cost and power consumption. Melanoma is the most aggressive form of skin cancer that increases the mortality rate. We aim to develop an optimized embedded SVM classifier dedicated for a low-cost handheld device for early detection of melanoma at the primary healthcare. In this paper, we propose a hardware/software co-design for implementing the SVM classifier onto FPGA to realize melanoma detection on a chip. The implemented SVM on a recent hybrid FPGA (Zynq) platform utilizing the modern UltraFast High-Level Synthesis design methodology achieves efficient melanoma classification on chip. The hardware implementation results demonstrate classification accuracy of 97.9%, and a significant hardware acceleration rate of 21 with only 3% resources utilization and 1.69W for power consumption. These results show that the implemented system on chip meets crucial embedded system constraints of high performance and low resources utilization, power consumption, and cost, while achieving efficient classification with high classification accuracy.

arXiv:2108.00570

Accelerating Markov Random Field Inference with Uncertainty Quantification

Statistical machine learning has widespread application in various domains. These methods include probabilistic algorithms, such as Markov Chain Monte-Carlo (MCMC), which rely on generating random numbers from probability distributions. These algorithms are computationally expensive on conventional processors, yet their statistical properties, namely interpretability and uncertainty quantification (UQ) compared to deep learning, make them an attractive alternative approach. Therefore, hardware specialization can be adopted to address the shortcomings of conventional processors in running these applications. In this paper, we propose a high-throughput accelerator for Markov Random Field (MRF) inference, a powerful model for representing a wide range of applications, using MCMC with Gibbs sampling. We propose a tiled architecture which takes advantage of near-memory computing, and memory optimizations tailored to the semantics of MRF. Additionally, we propose a novel hybrid on-chip/off-chip memory system and logging scheme to efficiently support UQ. This memory system design is not specific to MRF models and is applicable to applications using probabilistic algorithms. In addition, it dramatically reduces off-chip memory bandwidth requirements. We implemented an FPGA prototype of our proposed architecture using high-level synthesis tools and achieved 146MHz frequency for an accelerator with 32 function units on an Intel Arria 10 FPGA. Compared to prior work on FPGA, our accelerator achieves 26X speedup. Furthermore, our proposed memory system and logging scheme to support UQ reduces off-chip bandwidth by 71% for two applications. ASIC analysis in 15nm shows our design with 2048 function units running at 3GHz outperforms GPU implementations of motion estimation and stereo vision on Nvidia RTX2080Ti by 120X-210X, occupying only 7.7% of the area.

arXiv:2106.14089공식 코드 · Papers with Code

Accelerating Recurrent Neural Networks for Gravitational Wave Experiments

This paper presents novel reconfigurable architectures for reducing the latency of recurrent neural networks (RNNs) that are used for detecting gravitational waves. Gravitational interferometers such as the LIGO detectors capture cosmic events such as black hole mergers which happen at unknown times and of varying durations, producing time-series data. We have developed a new architecture capable of accelerating RNN inference for analyzing time-series data from LIGO detectors. This architecture is based on optimizing the initiation intervals (II) in a multi-layer LSTM (Long Short-Term Memory) network, by identifying appropriate reuse factors for each layer. A customizable template for this architecture has been designed, which enables the generation of low-latency FPGA designs with efficient resource utilization using high-level synthesis tools. The proposed approach has been evaluated based on two LSTM models, targeting a ZYNQ 7045 FPGA and a U250 FPGA. Experimental results show that with balanced II, the number of DSPs can be reduced up to 42% while achieving the same IIs. When compared to other FPGA-based LSTM designs, our design can achieve about 4.92 to 12.4 times lower latency.

arXiv:2105.01683

A reconfigurable neural network ASIC for detector front-end data compression at the HL-LHC

Despite advances in the programmable logic capabilities of modern trigger systems, a significant bottleneck remains in the amount of data to be transported from the detector to off-detector logic where trigger decisions are made. We demonstrate that a neural network autoencoder model can be implemented in a radiation tolerant ASIC to perform lossy data compression alleviating the data transmission problem while preserving critical information of the detector energy profile. For our application, we consider the high-granularity calorimeter from the CMS experiment at the CERN Large Hadron Collider. The advantage of the machine learning approach is in the flexibility and configurability of the algorithm. By changing the neural network weights, a unique data compression algorithm can be deployed for each sensor in different detector regions, and changing detector or collider conditions. To meet area, performance, and power constraints, we perform a quantization-aware training to create an optimized neural network hardware implementation. The design is achieved through the use of high-level synthesis tools and the hls4ml framework, and was processed through synthesis and physical layout flows based on a LP CMOS 65 nm technology node. The flow anticipates 200 Mrad of ionizing radiation to select gates, and reports a total area of 3.6 mm^2 and consumes 95 mW of power. The simulated energy consumption per inference is 2.4 nJ. This is the first radiation tolerant on-detector ASIC implementation of a neural network that has been designed for particle physics applications.

arXiv:2102.08138

IronMan: GNN-assisted Design Space Exploration in High-Level Synthesis via Reinforcement Learning

Despite the great success of High-Level Synthesis (HLS) tools, we observe several unresolved challenges: 1) the high-level abstraction of programming styles in HLS sometimes conceals optimization opportunities; 2) existing HLS tools do not provide flexible trade-off (Pareto) solutions among different objectives and constraints; 3) the actual quality of the resulting RTL designs is hard to predict. To address these challenges, we propose an end-to-end framework, namelyIronMan. The primary goal is to enable a flexible and automated design space exploration (DSE), to provide either optimal solutions under user-specified constraints, or various trade-offs among different objectives (such as different types of resources, area, and latency). Such DSE either requires tedious manual efforts or is not achievable to attain these goals through existing HLS tools. There are three components in IronMan: 1) GPP, a highly accurate graph-neural-network-based performance and resource predictor; 2) RLMD, a reinforcement-learning-based multi-objective DSE engine that explores the optimal resource allocation strategy, to provide Pareto solutions between different objectives; 3) CT, a code transformer to assist RLMD and GPP, which extracts the data flow graph from original HLS C/C++ and automatically generates synthesizable code with HLS directives. The experimental results show that: 1) GPP achieves high prediction accuracy, reducing prediction errors of HLS tools by 10.9x in resource utilization and 5.7x in timing; 2) RLMD obtains optimal or Pareto solutions that outperform the genetic algorithm and simulated annealing by 12.7% and 12.9%, respectively; 3) IronMan is able to find optimized solutions perfectly matching various DSP constraints, with 2.54x fewer DSPs and up to 6x shorter latency than those of HLS tools while being up to 400x faster than the heuristic algorithms and HLS tools.

arXiv:2012.01563

Accelerated Charged Particle Tracking with Graph Neural Networks on FPGAs

We develop and study FPGA implementations of algorithms for charged particle tracking based on graph neural networks. The two complementary FPGA designs are based on OpenCL, a framework for writing programs that execute across heterogeneous platforms, and hls4ml, a high-level-synthesis-based compiler for neural network to firmware conversion. We evaluate and compare the resource usage, latency, and tracking performance of our implementations based on a benchmark dataset. We find a considerable speedup over CPU-based execution is possible, potentially enabling such algorithms to be used effectively in future computing workflows and the FPGA-based Level-1 trigger at the CERN Large Hadron Collider.

arXiv:2007.10560

FPGA-Based Hardware Accelerator of Homomorphic Encryption for Efficient Federated Learning

With the increasing awareness of privacy protection and data fragmentation problem, federated learning has been emerging as a new paradigm of machine learning. Federated learning tends to utilize various privacy preserving mechanisms to protect the transferred intermediate data, among which homomorphic encryption strikes a balance between security and ease of utilization. However, the complicated operations and large operands impose significant overhead on federated learning. Maintaining accuracy and security more efficiently has been a key problem of federated learning. In this work, we investigate a hardware solution, and design an FPGA-based homomorphic encryption framework, aiming to accelerate the training phase in federated learning. The root complexity lies in searching for a compact architecture for the core operation of homomorphic encryption, to suit the requirement of federated learning about high encryption throughput and flexibility of configuration. Our framework implements the representative Paillier homomorphic cryptosystem with high level synthesis for flexibility and portability, with careful optimization on the modular multiplication operation in terms of processing clock cycle, resource usage and clock frequency. Our accelerator achieves a near-optimal execution clock cycle, with a better DSP-efficiency than existing designs, and reduces the encryption time by up to 71% during training process of various federated learning models.

arXiv:2005.06892

ZynqNet: An FPGA-Accelerated Embedded Convolutional Neural Network

Image Understanding is becoming a vital feature in ever more applications ranging from medical diagnostics to autonomous vehicles. Many applications demand for embedded solutions that integrate into existing systems with tight real-time and power constraints. Convolutional Neural Networks (CNNs) presently achieve record-breaking accuracies in all image understanding benchmarks, but have a very high computational complexity. Embedded CNNs thus call for small and efficient, yet very powerful computing platforms. This master thesis explores the potential of FPGA-based CNN acceleration and demonstrates a fully functional proof-of-concept CNN implementation on a Zynq System-on-Chip. The ZynqNet Embedded CNN is designed for image classification on ImageNet and consists of ZynqNet CNN, an optimized and customized CNN topology, and the ZynqNet FPGA Accelerator, an FPGA-based architecture for its evaluation. ZynqNet CNN is a highly efficient CNN topology. Detailed analysis and optimization of prior topologies using the custom-designed Netscope CNN Analyzer have enabled a CNN with 84.5% top-5 accuracy at a computational complexity of only 530 million multiplyaccumulate operations. The topology is highly regular and consists exclusively of convolutional layers, ReLU nonlinearities and one global pooling layer. The CNN fits ideally onto the FPGA accelerator. The ZynqNet FPGA Accelerator allows an efficient evaluation of ZynqNet CNN. It accelerates the full network based on a nested-loop algorithm which minimizes the number of arithmetic operations and memory accesses. The FPGA accelerator has been synthesized using High-Level Synthesis for the Xilinx Zynq XC-7Z045, and reaches a clock frequency of 200MHz with a device utilization of 80% to 90 %.

arXiv:2004.13075

A scalable and efficient convolutional neural network accelerator using HLS for a System on Chip design

This paper presents a configurable Convolutional Neural Network Accelerator (CNNA) for a System on Chip design (SoC). The goal was to accelerate inference of different deep learning networks on an embedded SoC platform. The presented CNNA has a scalable architecture which uses High Level Synthesis (HLS) and SystemC for the hardware accelerator. It is able to accelerate any Convolutional Neural Network (CNN) exported from Python and supports a combination of convolutional, max-pooling, and fully connected layers. A training method with fixed-point quantized weights is proposed and presented in the paper. The CNNA is template-based, enabling it to scale for different targets of the Xilinx Zynq platform. This approach enables design space exploration, which makes it possible to explore several configurations of the CNNA during C- and RTL-simulation, fitting it to the desired platform and model. The CNN VGG16 was used to test the solution on a Xilinx Ultra96 board using PYNQ. The result gave a high level of accuracy in training with an auto-scaled fixed-point Q2.14 format compared to a similar floating-point model. It was able to perform inference in 2.0 seconds, while having an average power consumption of 2.63 W, which corresponds to a power efficiency of 6.0 GOPS/W.

arXiv:2003.00671공식 코드 · Papers with Code

AutoPhase: Juggling HLS Phase Orderings in Random Forests with Deep Reinforcement Learning

The performance of the code a compiler generates depends on the order in which it applies the optimization passes. Choosing a good order--often referred to as the phase-ordering problem, is an NP-hard problem. As a result, existing solutions rely on a variety of heuristics. In this paper, we evaluate a new technique to address the phase-ordering problem: deep reinforcement learning. To this end, we implement AutoPhase: a framework that takes a program and uses deep reinforcement learning to find a sequence of compilation passes that minimizes its execution time. Without loss of generality, we construct this framework in the context of the LLVM compiler toolchain and target high-level synthesis programs. We use random forests to quantify the correlation between the effectiveness of a given pass and the program's features. This helps us reduce the search space by avoiding phase orderings that are unlikely to improve the performance of a given program. We compare the performance of AutoPhase to state-of-the-art algorithms that address the phase-ordering problem. In our evaluation, we show that AutoPhase improves circuit performance by 28% when compared to using the -O3 compiler flag, and achieves competitive results compared to the state-of-the-art solutions, while requiring fewer samples. Furthermore, unlike existing state-of-the-art solutions, our deep reinforcement learning solution shows promising result in generalizing to real benchmarks and 12,874 different randomly generated programs, after training on a hundred randomly generated programs.

arXiv:1806.09679

On the Resilience of RTL NN Accelerators: Fault Characterization and Mitigation

Machine Learning (ML) is making a strong resurgence in tune with the massive generation of unstructured data which in turn requires massive computational resources. Due to the inherently compute- and power-intensive structure of Neural Networks (NNs), hardware accelerators emerge as a promising solution. However, with technology node scaling below 10nm, hardware accelerators become more susceptible to faults, which in turn can impact the NN accuracy. In this paper, we study the resilience aspects of Register-Transfer Level (RTL) model of NN accelerators, in particular, fault characterization and mitigation. By following a High-Level Synthesis (HLS) approach, first, we characterize the vulnerability of various components of RTL NN. We observed that the severity of faults depends on both i) application-level specifications, i.e., NN data (inputs, weights, or intermediate), NN layers, and NN activation functions, and ii) architectural-level specifications, i.e., data representation model and the parallelism degree of the underlying accelerator. Second, motivated by characterization results, we present a low-overhead fault mitigation technique that can efficiently correct bit flips, by 47.3% better than state-of-the-art methods.

arXiv:1907.12952

Pyramid: Machine Learning Framework to Estimate the Optimal Timing and Resource Usage of a High-Level Synthesis Design

The emergence of High-Level Synthesis (HLS) tools shifted the paradigm of hardware design by making the process of mapping high-level programming languages to hardware design such as C to VHDL/Verilog feasible. HLS tools offer a plethora of techniques to optimize designs for both area and performance, but resource usage and timing reports of HLS tools mostly deviate from the post-implementation results. In addition, to evaluate a hardware design performance, it is critical to determine the maximum achievable clock frequency. Obtaining such information using static timing analysis provided by CAD tools is difficult, due to the multitude of tool options. Moreover, a binary search to find the maximum frequency is tedious, time-consuming, and often does not obtain the optimal result. To address these challenges, we propose a framework, called Pyramid, that uses machine learning to accurately estimate the optimal performance and resource utilization of an HLS design. For this purpose, we first create a database of C-to-FPGA results from a diverse set of benchmarks. To find the achievable maximum clock frequency, we use Minerva, which is an automated hardware optimization tool. Minerva determines the close-to-optimal settings of tools, using static timing analysis and a heuristic algorithm, and targets either optimal throughput or throughput-to-area. Pyramid uses the database to train an ensemble machine learning model to map the HLS-reported features to the results of Minerva. To this end, Pyramid re-calibrates the results of HLS to bridge the accuracy gap and enable developers to estimate the throughput or throughput-to-area of hardware design with more than 95% accuracy and alleviates the need to perform actual implementation for estimation.

arXiv:1905.03852

Machine Learning Based Routing Congestion Prediction in FPGA High-Level Synthesis

High-level synthesis (HLS) shortens the development time of hardware designs and enables faster design space exploration at a higher abstraction level. Optimization of complex applications in HLS is challenging due to the effects of implementation issues such as routing congestion. Routing congestion estimation is absent or inaccurate in existing HLS design methods and tools. Early and accurate congestion estimation is of great benefit to guide the optimization in HLS and improve the efficiency of implementation. However, routability, a serious concern in FPGA designs, has been difficult to evaluate in HLS without analyzing post-implementation details after Place and Route. To this end, we propose a novel method to predict routing congestion in HLS using machine learning and map the expected congested regions in the design to the relevant high-level source code. This is greatly beneficial in early identification of routability oriented bottlenecks in the high-level source code without running time-consuming register-transfer level (RTL) implementation flow. Experiments demonstrate that our approach accurately estimates vertical and horizontal routing congestion with errors of 6.71% and 10.05% respectively. By presenting Face Detection application as a case study, we show that by discovering the bottlenecks in high-level source code, routing congestion can be easily and quickly resolved compared to the efforts involved in RTL implementation and design feedback.

arXiv:1902.03192

Software-Defined FPGA Accelerator Design for Mobile Deep Learning Applications

Recently, the field of deep learning has received great attention by the scientific community and it is used to provide improved solutions to many computer vision problems. Convolutional neural networks (CNNs) have been successfully used to attack problems such as object recognition, object detection, semantic segmentation, and scene understanding. The rapid development of deep learning goes hand by hand with the adaptation of GPUs for accelerating its processes, such as network training and inference. Even though FPGA design exists long before the use of GPUs for accelerating computations and despite the fact that high-level synthesis (HLS) tools are getting more attractive, the adaptation of FPGAs for deep learning research and application development is poor due to the requirement of hardware design related expertise. This work presents a workflow for deep learning mobile application acceleration on small low-cost low-power FPGA devices using HLS tools. This workflow eases the design of an improved version of the SqueezeJet accelerator used for the speedup of mobile-friendly low-parameter ImageNet class CNNs, such as the SqueezeNet v1.1 and the ZynqNet. Additionally, the workflow includes the development of an HLS-driven analytical model which is used for performance estimation of the accelerator. This model can be also used to direct the design process and lead to future design improvements and optimizations.