1 December 2019 in IEEE Journal on Emerging and Selected Topics in Circuits and Systems
1 January 2021 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
In order to effectively reduce buffer energy consumption, which constitutes a significant part of the total energy consumption in a convolutional neural network (CNN), it is useful to apply different amounts of energy conservation effort to the different levels of a CNN as the buffer energy to total energy usage ratios can differ quite substantially across the layers of a CNN. This paper proposes layerwise buffer voltage scaling as an effective technique for reducing buffer access energy. Error-resilience analysis, including inter-layer effects, conducted during design-time is used to determine the specific buffer supply voltage to be used for each layer of a CNN. Then these layer-specific buffer supply voltages are used in the CNN for image classification inference. Error injection experiments with three different types of CNN architectures show that, with this technique, the buffer access energy and overall system energy can be reduced by up to 68.41% and 33.68%, respectively, without sacrificing image classification accuracy.
6 September 2019 in IEEE Access
Neural networks trained using images with a certain type of distortion should be better at classifying test images with the same type of distortion than generally-trained neural networks, given other factors being equal. Based on this observation, an ensemble of convolutional neural networks (CNNs) trained with different types and degrees of distortions is used. However, instead of simply classifying test images of unknown distortion types with the entire ensemble of CNNs, an extra tiny CNN is specifically trained to distinguish between the different types and degrees of distortions. Then, only the dedicated CNN for that specific type and degree of distortion, as determined by the tiny CNN, is activated and used to classify a possibly distorted test image. This proposed architecture, referred to as a selective deep convolutional neural network (DCNN), is implemented and found to result in high accuracy with low hardware costs. Detailed simulations with realistic image distortion scenarios using three popular datasets show that memory, MAC operations, and energy savings of up to 93.68%, 93.61%, and 91.92%, respectively, can be achieved with almost no reduction in image classification accuracy. The proposed selective DCNN scores up to $2.18\times $ higher than the state-of-the-art DCNN model when evaluated using NetScore, a comprehensive metric that considers both CNN performance and hardware cost. In addition, it is shown that even higher hardware cost reduction can be achieved when selective DCNN is combined with previously proposed model compression techniques. Finally, experiments conducted with extended types and degrees of image distortion show that selective DCNN is highly scalable.
9 December 2024 in IEEE Sensors Journal
Multi-touch detection algorithms are crucial for precise interaction with touch screens. However, as existing multi-touch detection algorithms only target the Good Ground Mass (GGM) environment, performance drops sharply when the ground state is unstable, such as the Low Ground Mass (LGM) environment. This paper introduces an enhanced multi-touch detection algorithm tailored for both ground environments, addressing deficiencies in recognizing large touch areas and precise coordinates. First, the Center of Thumb Search (CTS) method combined with an adaptive Valley Point Division (VPD) skip process for large circular touches, such as thumb touches, enables detection without excessive segmentation. Secondly, conditional VPD thresholding is designed to distinguish similar single and multi-touch in LGM environment. This algorithm posed challenges that negatively impacted the detection performance in the GGM environment; however, these issues were addressed through the development of Ground State Classifier (GSC). Lastly, CTS algorithm facilitates the distinction of the center of a large-sized thumb touch, enhancing the resolution in closely spaced touch scenarios by properly partitioning touch groups. Experimental results demonstrate significant improvements in accuracy and linearity. We have quantitatively confirmed substantial enhancements in performance from a user standpoint, achieving 96.07% in accuracy for the total dataset and ×13.36 better in linearity. These innovations collectively advance the state of touch detection technology in challenging LGM environments, presenting a robust framework for future applications.
17 May 2022 in IEEE Transactions on Circuits and Systems I: Regular Papers
This paper presents an advanced offline scheduling scheme to improve the accelerator efficiency, especially for the highly-pruned convolutional neural networks (HP-CNNs). Based on the existing outlier-aware accelerator design, we demonstrate the efficiency drop of HP-CNN processing for the first time, and element-wise channel merging is proposed to make a dense processing sequence even for the highly-pruned model. The dedicated hardware architecture is also presented to process the merged channels with the minimum hardware-level overheads, improving the energy efficiency for handling HP-CNNs by preserving the hardware utilization. We further investigate the optimal size of accumulator and multiplexer, in addition to the number of merged channels, exploiting the attractive energy-performance trade-offs. As a result, unlike the practical HP-CNNs for the on-device solutions, the proposed method enhances the overall efficiency by up to 33% compared to the state-of-the-art schemes.