Reducing MAC operation in convolutional neural network with sign prediction

Chang, Jung Min; Choi, Yongjun; Lee, Taegyoung; Cho, Junhee

doi:10.1109/ictc.2018.8539530

Cited by 15 publications

(5 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, in the HW-oriented approach (Figure 4.b), the developers are mainly focusing on designing enhanced hardware platforms that are optimized for embedded applications in order to run current and future state-of-the-art ML algorithms. This often involves investigating the bottlenecks in an existing architecture with regard to computations within a ML framework, like neural networks, and the design of hardware accelerator modules to improve throughput and consumption: e.g., reducing computational complexity in convolution layers [49], [50], efficient, low-power and feature-rich perceptrons [51], enhanced data caches [52]. In other cases, the developers design new hardware platforms optimized for embedded applications with extended digital signal processing capabilities already integrated [53].…”

Section: Tinyml Workflowsmentioning

confidence: 99%

A Machine Learning-Oriented Survey on Tiny Machine Learning

Capogrosso,

Cunico,

Cheng

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The emergence of Tiny Machine Learning (TinyML) has positively revolutionized the field of Artificial Intelligence by promoting the joint design of resource-constrained IoT hardware devices and their learning-based software architectures. TinyML carries an essential role within the fourth and fifth industrial revolutions in helping societies, economies, and individuals employ effective AI-infused computing technologies (e.g., smart cities, automotive, and medical robotics). Given its multidisciplinary nature, the field of TinyML has been approached from many different angles: this comprehensive survey wishes to provide an up-to-date overview focused on all the learning algorithms within TinyML-based solutions. The survey is based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodological flow, allowing for a systematic and complete literature survey. In particular, firstly, we will examine the three different workflows for implementing a TinyML-based system, i.e., MLoriented, HW-oriented, and co-design. Secondly, we propose a taxonomy that covers the learning panorama under the TinyML lens, examining in detail the different families of model optimization and design, as well as the state-of-the-art learning techniques. Thirdly, this survey will present the distinct features of hardware devices and software tools that represent the current state-of-the-art for TinyML intelligent edge applications. Finally, we discuss the challenges and future directions.

show abstract

Section: Tinyml Workflowsmentioning

confidence: 99%

A Machine Learning-Oriented Survey on Tiny Machine Learning

Capogrosso,

Cunico,

Cheng

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…As compare to the single stream, our model has added some additional cost, we show that the existing architecture is highly able to augment multiple feature spaces using single stream. We calculate and compare the number of parameters and computational complexity (MACs) [74] of HAFS with compare to the existing ResNext101 and BESS in Fig. 11.…”

Section: Complexity Analysis Of Bess and Hafsmentioning

confidence: 99%

“…In the case of single stream HAFS has increased the number of parameters which is less and comparable. Comparative computa-tional complexity is measure using the popular metric called "Multiply-and-accumulates (MACs) per frames" [74]. We notice that HAFS has nearly similar MACs (38.58GM ACs) with compare to the existing ResNext101(Basic*) and BESS (38.57GM ACs).…”

Section: Complexity Analysis Of Bess and Hafsmentioning

confidence: 99%

Batch Entropy Supervised Convolutional Neural Networks for Feature Extraction and Harmonizing for Action Recognition

et al. 2020

View full text Add to dashboard Cite

“…Akhlaghi et al [1] predict during the convolution computation whether the convolution results will end up negative. Song et al [42], Lin et al [30], and Chang et al [4] predict whether an entire convolution result is negative according to a partial result yielded by the input MSB bits. Huan et al [19] avoid convolution multiplications by predicting and skipping near-zero valued data, given certain thresholds.…”

Section: Related Workmentioning

confidence: 99%

Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks

Shomron

Banner

Shkolnik

et al. 2019

Preprint

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) introduce stateof-the-art results for various tasks with the price of high computational demands. Inspired by the observation that spatial correlation exists in CNN output feature maps (ofms), we propose a method to dynamically predict whether ofm activations are zero-valued or not according to their neighboring activation values, thereby avoiding zerovalued activations and reducing the number of convolution operations. We implement the zero activation predictor (ZAP) with a lightweight CNN, which imposes negligible overheads and is easy to train and deploy on existing models. Furthermore, without model retraining, the same ZAP can be tuned to many different operating points along the accuracy-savings trade-off curve. For example, using VGG-16 and the ILSVRC-2012 dataset, two different operating points achieve a reduction of 20% and 30% multiplyaccumulate (MAC) operations with top-1/top-5 accuracy degradation of 0.1%/0.04% and 1.3%/0.7% without finetuning of the entire model, respectively. Considering oneepoch fine-tuning, 45% MAC operations may be reduced with 1.3%/0.7% accuracy degradation.

show abstract

Reducing MAC operation in convolutional neural network with sign prediction

Cited by 15 publications

References 7 publications

A Machine Learning-Oriented Survey on Tiny Machine Learning

A Machine Learning-Oriented Survey on Tiny Machine Learning

Batch Entropy Supervised Convolutional Neural Networks for Feature Extraction and Harmonizing for Action Recognition

Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks

Contact Info

Product

Resources

About