Search

We introduce an area/energy-efficient precision-scalable neural network accelerator architecture. Previous precision-scalable hardware accelerators have limitations such as the under-utilization of multipliers for low bit-width operations and the large area overhead to support various bit precisions. To mitigate the problems, we first propose a bitwise summation, which reduces the area overhead for the bit-width scaling. In addition, we present a channel-wise aligning scheme (CAS) to efficiently fetch inputs and weights from on-chip SRAM buffers and a channel-first and pixel-last tiling (CFPL) scheme to maximize the utilization of multipliers on various kernel sizes. A test chip was implemented in 28-nm CMOS technology, and the experimental results show that the throughput and energy efficiency of our chip are up to 7.7 $\times $ and 1.64 $\times $ higher than those of the state-of-the-art designs, respectively. Moreover, additional 1.5–3.4 $\times $ throughput gains can be achieved using the CFPL method compared to the CAS. [ABSTRACT FROM AUTHOR]

This paper proposes the optimal white point settings of consumer televisions observed in different viewing conditions. The ratings from visual experiments were modeled with a bivariate Gaussian distribution to predict the optimal white points. Under dark surround conditions, where there is no effect of lighting chromaticity, the averaged optimal white point was located at a correlated color temperature of 7346 K and a Duv of −0.0015. The optimal white point was shifted toward a higher color temperature compared to the current standards. The results also indicated that current white point settings of major television manufacturers are too bluish. Neither televisions that meet the current standards nor televisions in the market would appear as true white under the viewing conditions employed in this paper. Moreover, the optimal white point depended largely on viewing conditions. Image content and ambient lighting conditions were observed to be important factors that influence the perception of the white point. The research therefore proposed to automatically change the white point of the display based on the image content and the real-time measurements of the surrounding ambient light falling on the screen. Ultimately, this paper might be of interest to the industry as a basis for future deliberations on target white points for consumer televisions designed to provide a better color appearance. [ABSTRACT FROM AUTHOR]

In this letter, we demonstrate the conductive-bridging RAM (CBRAM) with excellent multi-level cell (MLC) and linear conductance characteristics for an artificial synaptic device of neuromorphic systems. Our findings show that inherent characteristics of CBRAM can achieve the linear conductance and MLC characteristics as a product of an integer unit of the conductance. However, uncontrolled metal-ion injection into the switching layer results in a significant degradation of device uniformity, leading to degradation in the classification accuracy. Thus, we introduce a multi-layer CBRAM configuration (Cu/HfO2/Ta/Cu2S/W) to control the ionic motion in electrolytes. As a result of device engineering, highly improved classification accuracy is achieved using CIFAR-10 data set. [ABSTRACT FROM PUBLISHER]

Artificial neural network (ANN) computations based on graphics processing units (GPUs) consume high power. Resistive random-access memory (RRAM) has been gaining attention as a promising technology for implementing power-efficient ANNs, replacing GPU. However, nonlinear I – V characteristics of RRAM devices have been limiting its use for ANN implementation. In this letter, we propose a method and a circuit to address issues due to the nonlinear I – V characteristics. We demonstrate the feasibility of the method by simulating its application to multiple neural networks, from multi-layer perceptron to deep convolutional neural network based on a typical RRAM model. Results from classifying datasets including ImageNet show that the proposed method produces much higher accuracy than the naive linear mapping for a wide range of nonlinearity. [ABSTRACT FROM PUBLISHER]