An official website of the United States government
Large-scale deep neural networks (DNNs) are both compute and memory intensive. As the size of DNNs continues to grow, it is critical to improve the energy efficiency and performance while maintaining accuracy. For DNNs, the model size is an important factor affecting performance, scalability and energy efficiency. Weight pruning achieves good compression ratios but suffers from three drawbacks: 1) the irregular network structure after pruning, which affects performance and throughput; 2) the increased training complexity; and 3) the lack of rigirous guarantee of compression ratio and inference accuracy. To overcome these limitations, this paper proposes CirCNN, a principled approach to represent weights and process neural networks using block-circulant matrices. CirCNN utilizes the Fast Fourier Transform (FFT)-based fast multiplication, simultaneously reducing the computational complexity (both in inference and training) from O(n2) to O(n log n) and the storage complexity from O(n2) to O(n), with negligible accuracy loss. Compared to other approaches, CirCNN is distinct due to its mathematical rigor: the DNNs based on CirCNN can converge to the same "effectiveness" as DNNs without compression. We propose the CirCNN architecture, a universal DNN inference engine that can be implemented in various hardware/software platforms with configurable network architecture (e.g., layer type, size, scales, etc.). In CirCNN architecture: 1) Due to the recursive property, FFT can be used as the key computing kernel, which ensures universal and small-footprint implementations. 2) The compressed but regular network structure avoids the pitfalls of the network pruning and facilitates high performance and throughput with highly pipelined and parallel design. To demonstrate the performance and energy efficiency, we test CirCNN in FPGA, ASIC and embedded processors. Our results show that CirCNN architecture achieves very high energy efficiency and performance with a small hardware footprint. Based on the FPGA implementation and ASIC synthesis results, CirCNN achieves 6 - 102X energy efficiency improvements compared with the best state-of-the-art results.
more »
« less
Towards Ultra-High Performance and Energy Efficiency of Deep Learning Systems: An Algorithm-Hardware Co-Optimization Framework
Hardware accelerations of deep learning systems have been extensively investigated in industry and academia. The aim of this paper is to achieve ultra-high energy efficiency and performance for hardware implementations of deep neural networks (DNNs). An algorithm-hardware co-optimization framework is developed, which is applicable to different DNN types, sizes, and application scenarios. The algorithm part adopts the general block-circulant matrices to achieve a fine-grained tradeoff of accuracy and compression ratio. It applies to both fully-connected and convolutional layers and contains a mathematically rigorous proof of the effectiveness of the method. The proposed algorithm reduces computational complexity per layer from O(n2 ) to O(n log n) and storage complexity from O(n2) to O(n), both for training and inference. The hardware part consists of highly efficient Field Programmable Gate Array (FPGA)-based implementations using effective reconfiguration, batch processing, deep pipelining, resource re-using, and hierarchical control. Experimental results demonstrate that the proposed framework achieves at least 152X speedup and 71X energy efficiency gain compared with IBM TrueNorth processor under the same test accuracy. It achieves at least 31X energy efficiency gain compared with the reference FPGA-based work.
more »
« less
- Award ID(s):
- 1733701
- PAR ID:
- 10066606
- Date Published:
- Journal Name:
- The Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18)
- Page Range / eLocation ID:
- 4235-4243
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Deep neural networks (DNNs) have emerged as the most powerful machine learning technique in numerous artificial intelligent applications. However, the large sizes of DNNs make themselves both computation and memory intensive, thereby limiting the hardware performance of dedicated DNN accelerators. In this paper, we propose a holistic framework for energy-efficient high-performance highly-compressed DNN hardware design. First, we propose block-circulant matrix-based DNN training and inference schemes, which theoretically guarantee Big-O complexity reduction in both computational cost (from O(n2) to O(n log n)) and storage requirement (from O(n2) to O(n)) of DNNs. Second, we dedicatedly optimize the hardware architecture, especially on the key fast Fourier transform (FFT) module, to improve the overall performance in terms of energy efficiency, computation performance and resource cost. Third, we propose a design flow to perform hardware-software co-optimization with the purpose of achieving good balance between test accuracy and hardware performance of DNNs. Based on the proposed design flow, two block-circulant matrix-based DNNs on two different datasets are implemented and evaluated on FPGA. The fixed-point quantization and the proposed block-circulant matrix-based inference scheme enables the network to achieve as high as 3.5 TOPS computation performance and 3.69 TOPS/W energy efficiency while the memory is saved by 108X ~ 116X with negligible accuracy degradation.more » « less
-
Recurrent Neural Networks (RNNs) are becoming increasingly important for time series-related applications which require efficient and real-time implementations. The two major types are Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) networks. It is a challenging task to have real-time, efficient, and accurate hardware RNN implementations because of the high sensitivity to imprecision accumulation and the requirement of special activation function implementations. Recently two works have focused on FPGA implementation of inference phase of LSTM RNNs with model compression. First, ESE uses a weight pruning based compressed RNN model but suffers from irregular network structure after pruning. The second work C-LSTM mitigates the irregular network limitation by incorporating block-circulant matrices for weight matrix representation in RNNs, thereby achieving simultaneous model compression and acceleration. A key limitation of the prior works is the lack of a systematic design optimization framework of RNN model and hardware implementations, especially when the block size (or compression ratio) should be jointly optimized with RNN type, layer size, etc. In this paper, we adopt the block-circulant matrixbased framework, and present the Efficient RNN (E-RNN) framework for FPGA implementations of the Automatic Speech Recognition (ASR) application. The overall goal is to improve performance/energy efficiency under accuracy requirement. We use the alternating direction method of multipliers (ADMM) technique for more accurate block-circulant training, and present two design explorations providing guidance on block size and reducing RNN training trials. Based on the two observations, we decompose E-RNN in two phases: Phase I on determining RNN model to reduce computation and storage subject to accuracy requirement, and Phase II on hardware implementations given RNN model, including processing element design/optimization, quantization, activation implementation, etc. 1 Experimental results on actual FPGA deployments show that E-RNN achieves a maximum energy efficiency improvement of 37.4× compared with ESE, and more than 2× compared with C-LSTM, under the same accuracy.more » « less
-
Algorithm-hardware Co-optimization for Energy-efficient Drone Detection on Resource-constrained FPGAConvolutional neural network (CNN)-based object detection has achieved very high accuracy; e.g., single-shot multi-box detectors (SSDs) can efficiently detect and localize various objects in an input image. However, they require a high amount of computation and memory storage, which makes it difficult to perform efficient inference on resource-constrained hardware devices such as drones or unmanned aerial vehicles (UAVs). Drone/UAV detection is an important task for applications including surveillance, defense, and multi-drone self-localization and formation control. In this article, we designed and co-optimized an algorithm and hardware for energy-efficient drone detection on resource-constrained FPGA devices. We trained an SSD object detection algorithm with a custom drone dataset. For inference, we employed low-precision quantization and adapted the width of the SSD CNN model. To improve throughput, we use dual-data rate operations for DSPs to effectively double the throughput with limited DSP counts. For different SSD algorithm models, we analyze accuracy or mean average precision (mAP) and evaluate the corresponding FPGA hardware utilization, DRAM communication, and throughput optimization. We evaluated the FPGA hardware for a custom drone dataset, Pascal VOC, and COCO2017. Our proposed design achieves a high mAP of 88.42% on the multi-drone dataset, with a high energy efficiency of 79 GOPS/W and throughput of 158 GOPS using the Xilinx Zynq ZU3EG FPGA device on the Open Vision Computer version 3 (OVC3) platform. Our design achieves 1.1 to 8.7× higher energy efficiency than prior works that used the same Pascal VOC dataset, using the same FPGA device, but at a low-power consumption of 2.54 W. For the COCO dataset, our MobileNet-V1 implementation achieved an mAP of 16.8, and 4.9 FPS/W for energy-efficiency, which is ∼ 1.9× higher than prior FPGA works or other commercial hardware platforms.more » « less
-
Quantization-Based Optimization Algorithm for Hardware Implementation of Convolution Neural NetworksConvolutional neural networks (CNNs) have demonstrated remarkable performance in many areas but require significant computation and storage resources. Quantization is an effective method to reduce CNN complexity and implementation. The main research objective is to develop a scalable quantization algorithm for CNN hardware design and model the performance metrics for the purpose of CNN implementation in resource-constrained devices (RCDs) and optimizing layers in deep neural networks (DNNs). The algorithm novelty is based on blending two quantization techniques to perform full model quantization with optimum accuracy, and without additional neurons. The algorithm is applied to a selected CNN model and implemented on an FPGA. Implementing CNN using broad data is not possible due to capacity issues. With the proposed quantization algorithm, we succeeded in implementing the model on the FPGA using 16-, 12-, and 8-bit quantization. Compared to the 16-bit design, the 8-bit design offers a 44% decrease in resource utilization, and achieves power and energy reductions of 41% and 42%, respectively. Models show that trading off one quantization bit yields savings of approximately 5.4K LUTs, 4% logic utilization, 46.9 mW power, and 147 μJ energy. The models were also used to estimate performance metrics for a sample DNN design.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
