Accuracy vs. Efficiency: Achieving both Through Hardware-Aware Quantization and Reconfigurable Architecture with Mixed Precision

Chang, Libo; Zhang, Shengbing; Du, Heng; Wang, Shiyu; Qiu, Meikang; Wang, Jihe

doi:10.1109/ispa-bdcloud-socialcom-sustaincom52081.2021.00033

Cited by 2 publications

(3 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FBNet [19] HotNAS [16] Analytical estimation Compute a rough estimate using the processing time, the stall time and the starting time Latency FNAS [34] NASCaps [26] Energy NASCaps [26] Predictive model Build a ML model to predict the cost using architecture and dataset features Latency proxylessNAS [14] NASAIC [38] developer. For example, model prediction methods require expert knowledge to select the best features and verify them, lookup table methods also take a lot of time to build code compilation for each operation on the target hardware.…”

Section: Latencymentioning

confidence: 99%

“…The hardware search space can be further classified as follows: Parameter configuration : The hardware search space is formalised by a group of parameter configurations. The Hardware Search Space of FNAS [34] consists of four tiled parameters of convolution. It can search for the best performing model on a given dataset and the optimised parameters needed to deploy it in an FPGA chip for deep learning.…”

Section: Search Space Of Nas On Hardware Devicesmentioning

confidence: 99%

See 1 more Smart Citation

Neural architecture search for resource constrained hardware devices: A survey

Yang

Zhan

Jiang

et al. 2023

IET Cyber-Phy Sys Theory & Ap

View full text Add to dashboard Cite

With the emergence of powerful and low‐energy Internet of Things devices, deep learning computing is increasingly applied to resource‐constrained edge devices. However, the mismatch between hardware devices with low computing capacity and the increasing complexity of Deep Neural Network models, as well as the growing real‐time requirements, bring challenges to the design and deployment of deep learning models. For example, autonomous driving technologies rely on real‐time object detection of the environment, which cannot tolerate the extra latency of sending data to the cloud, processing and then sending the results back to edge devices. Many studies aim to find innovative ways to reduce the size of deep learning models, the number of Floating‐point Operations per Second, and the time overhead of inference. Neural Architecture Search (NAS) makes it possible to automatically generate efficient neural network models. The authors summarise the existing NAS methods on resource‐constrained devices and categorise them according to single‐objective or multi‐objective optimisation. We review the search space, the search algorithm and the constraints of NAS on hardware devices. We also explore the challenges and open problems of hardware NAS.

show abstract

Section: Latencymentioning

confidence: 99%

Section: Search Space Of Nas On Hardware Devicesmentioning

confidence: 99%

Neural architecture search for resource constrained hardware devices: A survey

Yang

Zhan

Jiang

et al. 2023

IET Cyber-Phy Sys Theory & Ap

View full text Add to dashboard Cite

show abstract

“…Some researchers have embarked on the study of mixed-precision algorithms, which has led to many hardware accelerator designs. Chang et al ( 2021 ) designed a reconfigurable CNN processor, which can reconstruct the computing unit and the on-chip buffer according to the computing characteristics of the model with mixed-precision quantization. Jiang et al ( 2020 ) designed the PRArch accelerator architecture which support both conventional dense convolution and aggregated sparse convolution and implement mixed-precision convolution on fix-precision systolic arrays.…”

Section: Introductionmentioning

confidence: 99%

ALBSNN: ultra-low latency adaptive local binary spiking neural network with accuracy loss estimator

Pei,

Xu,

et al. 2023

Front. Neurosci.

View full text Add to dashboard Cite

Spiking neural network (SNN) is a brain-inspired model with more spatio-temporal information processing capacity and computational energy efficiency. However, with the increasing depth of SNNs, the memory problem caused by the weights of SNNs has gradually attracted attention. In this study, we propose an ultra-low latency adaptive local binary spiking neural network (ALBSNN) with accuracy loss estimators, which dynamically selects the network layers to be binarized to ensure a balance between quantization degree and classification accuracy by evaluating the error caused by the binarized weights during the network learning process. At the same time, to accelerate the training speed of the network, the global average pooling (GAP) layer is introduced to replace the fully connected layers by combining convolution and pooling. Finally, to further reduce the error caused by the binary weight, we propose binary weight optimization (BWO), which updates the overall weight by directly adjusting the binary weight. This method further reduces the loss of the network that reaches the training bottleneck. The combination of the above methods balances the network's quantization and recognition ability, enabling the network to maintain the recognition capability equivalent to the full precision network and reduce the storage space by more than 20%. So, SNNs can use a small number of time steps to obtain better recognition accuracy. In the extreme case of using only a one-time step, we still can achieve 93.39, 92.12, and 69.55% testing accuracy on three traditional static datasets, Fashion- MNIST, CIFAR-10, and CIFAR-100, respectively. At the same time, we evaluate our method on neuromorphic N-MNIST, CIFAR10-DVS, and IBM DVS128 Gesture datasets and achieve advanced accuracy in SNN with binary weights. Our network has greater advantages in terms of storage resources and training time.

show abstract

Accuracy vs. Efficiency: Achieving both Through Hardware-Aware Quantization and Reconfigurable Architecture with Mixed Precision

Cited by 2 publications

References 25 publications

Neural architecture search for resource constrained hardware devices: A survey

Neural architecture search for resource constrained hardware devices: A survey

ALBSNN: ultra-low latency adaptive local binary spiking neural network with accuracy loss estimator

Contact Info

Product

Resources

About