10 results on '"Ma, Kaisheng"'
Search Results
2. Design of Nonvolatile SRAM with Ferroelectric FETs for Energy-Efficient Backup and Restore.
- Author
-
Li, Xueqing, Ma, Kaisheng, George, Sumitha, Khwa, Win-San, Sampson, John, Gupta, Sumeet, Liu, Yongpan, Chang, Meng-Fan, Datta, Suman, and Narayanan, Vijaykrishnan
- Subjects
- *
FERROELECTRIC devices , *FIELD-effect transistors , *ENERGY consumption , *NONVOLATILE memory , *POWER resources - Abstract
Nonvolatile SRAM (nvSRAM) has emerged as a promising approach to reducing the standby energy consumption by storing the state into an in situ nonvolatile memory element and shutting down the power supply. Existing nvSRAM solutions based on a nonvolatile backup in magnetic tunnel junction and ReRAM, however, are costly in backup and restore energy due to static current. This cost results in a long break-even time (BET) when compared with a lowered voltage standby volatile SRAM. This brief proposes an nvSRAM based on ferroelectric FETs (FeFETs) that are capable of fully avoiding such static current. A simple differential backup and restore circuitry is proposed, achieving sub-fJ/cell total energy per backup and restore operation at the 10-nm node. This leads to hundreds of times BET improvement over existing ReRAM nvSRAM solutions. This nvSRAM also indicates the future FeFET design trends for such memory-logic synergy. [ABSTRACT FROM PUBLISHER]
- Published
- 2017
- Full Text
- View/download PDF
3. Nonvolatile Processor Architectures: Efficient, Reliable Progress with Unstable Power.
- Author
-
Ma, Kaisheng, Li, Xueqing, Swaminathan, Karthik, Zheng, Yang, Li, Shuangchen, Liu, Yongpan, Xie, Yuan, Sampson, John Jack, and Narayanan, Vijaykrishnan
- Subjects
- *
NONVOLATILE memory , *COMPUTER software , *COMPUTER architecture , *COMPUTER power supply , *POWER resources - Abstract
Nonvolatile processors (NVPs) have integrated nonvolatile memory to preserve task-intermediate on-chip state during power emergencies. NVPs hide data backup and restoration from the executing software to provide an execution mode that will always eventually complete the current task. NVPs are emerging as a promising solution for energy-harvesting scenarios, in which the available power supply is unstable and intermittent, because of their ability to ensure that even short periods of sufficient power, on the order of tens of instructions, will result in net forward progress. This article explores the design space for an NVP across different architectures, input power sources, and policies for maximizing forward progress in a framework calibrated using measured results from a fabricated NVP. The authors propose a heterogeneous microarchitecture solution that more efficiently capitalizes on ephemeral power surpluses. [ABSTRACT FROM PUBLISHER]
- Published
- 2016
- Full Text
- View/download PDF
4. Nonvolatile Processor Architecture Exploration for Energy-Harvesting Applications.
- Author
-
Ma, Kaisheng, Li, Xueqing, Li, Shuangchen, Liu, Yongpan, Sampson, John Jack, Xie, Yuan, and Narayanan, Vijaykrishnan
- Subjects
- *
MICROPROCESSOR design & construction , *ENERGY harvesting , *INTERNET of things , *NONVOLATILE memory , *ELECTRONIC circuits - Abstract
This article provides insights into the design of nonvolatile processors (NVPs) for batteryless applications in the Internet of Things (IoT), in which ambient energy-harvesting techniques provide the power. Achieving reliable, continuous, forward computation with an unstable, intermittent power supply motivates the transition from conventional volatile processors to emerging NVPs. The authors discuss the various design factors and tradeoffs involved in optimizing this forward progress. This article provides a guide for future IoT applications, revealing inherent features of energy-harvesting NVP design. [ABSTRACT FROM PUBLISHER]
- Published
- 2015
- Full Text
- View/download PDF
5. Self-Distillation: Towards Efficient and Compact Neural Networks.
- Author
-
Zhang, Linfeng, Bao, Chenglong, and Ma, Kaisheng
- Subjects
- *
ARTIFICIAL neural networks , *KNOWLEDGE transfer - Abstract
Remarkable achievements have been obtained by deep neural networks in the last several years. However, the breakthrough in neural networks accuracy is always accompanied by explosive growth of computation and parameters, which leads to a severe limitation of model deployment. In this paper, we propose a novel knowledge distillation technique named self-distillation to address this problem. Self-distillation attaches several attention modules and shallow classifiers at different depths of neural networks and distills knowledge from the deepest classifier to the shallower classifiers. Different from the conventional knowledge distillation methods where the knowledge of the teacher model is transferred to another student model, self-distillation can be considered as knowledge transfer in the same model - from the deeper layers to the shallow layers. Moreover, the additional classifiers in self-distillation allow the neural network to work in a dynamic manner, which leads to a much higher acceleration. Experiments demonstrate that self-distillation has consistent and significant effectiveness on various neural networks and datasets. On average, 3.49 and 2.32 percent accuracy boost are observed on CIFAR100 and ImageNet. Besides, experiments show that self-distillation can be combined with other model compression methods, including knowledge distillation, pruning and lightweight model design. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
6. Advancing Nonvolatile Computing With Nonvolatile NCFET Latches and Flip-Flops.
- Author
-
Li, Xueqing, George, Sumitha, Ma, Kaisheng, Tsai, Wei-Yu, Aziz, Ahmedullah, Sampson, John, Gupta, Sumeet Kumar, Chang, Meng-Fan, Liu, Yongpan, Datta, Suman, and Narayanan, Vijaykrishnan
- Subjects
- *
NONVOLATILE memory , *FLIP-flop circuits , *LOGIC circuits - Abstract
Nonvolatile computing has been proven to be effective in dealing with power supply outages for on-chip check-pointing in emerging energy-harvesting Internet-of-Things applications. It also plays an important role in power-gating to cut off leakage power for higher energy efficiency. However, existing on-chip state backup solutions for D flip–flop (DFF) have a bottleneck of significant energy and/or latency penalties which limit the overall energy efficiency and computing progress. Meanwhile, these solutions rely on external control that limits compatibility and increases system complexity. This paper proposes an approach to fundamentally advancing the nonvolatile computing paradigm by intrinsically nonvolatile area-efficient latches and flip–flops designs using negative capacitance FET. These designs consume fJ-level energy and ns-level intrinsic latency for a backup plus restore operation, e.g., 2.4 fJ in energy and 1.1 ns in time for one proposed nonvolatile DFF with a supply power of 0.80 V. [ABSTRACT FROM AUTHOR]
- Published
- 2017
- Full Text
- View/download PDF
7. Non-Structured DNN Weight Pruning—Is It Beneficial in Any Platform?
- Author
-
Ma, Xiaolong, Lin, Sheng, Ye, Shaokai, He, Zhezhi, Zhang, Linfeng, Yuan, Geng, Tan, Sia Huat, Li, Zhengang, Fan, Deliang, Qian, Xuehai, Lin, Xue, Ma, Kaisheng, and Wang, Yanzhi
- Subjects
- *
ARTIFICIAL neural networks , *DYNAMIC random access memory , *COMMUNITIES , *RANDOM access memory , *ENERGY consumption , *STATIC random access memory - Abstract
Large deep neural network (DNN) models pose the key challenge to energy efficiency due to the significantly higher energy consumption of off-chip DRAM accesses than arithmetic or SRAM operations. It motivates the intensive research on model compression with two main approaches. Weight pruning leverages the redundancy in the number of weights and can be performed in a non-structured, which has higher flexibility and pruning rate but incurs index accesses due to irregular weights, or structured manner, which preserves the full matrix structure with a lower pruning rate. Weight quantization leverages the redundancy in the number of bits in weights. Compared to pruning, quantization is much more hardware-friendly and has become a “must-do” step for FPGA and ASIC implementations. Thus, any evaluation of the effectiveness of pruning should be on top of quantization. The key open question is, with quantization, what kind of pruning (non-structured versus structured) is most beneficial? This question is fundamental because the answer will determine the design aspects that we should really focus on to avoid the diminishing return of certain optimizations. This article provides a definitive answer to the question for the first time. First, we build ADMM-NN-S by extending and enhancing ADMM-NN, a recently proposed joint weight pruning and quantization framework, with the algorithmic supports for structured pruning, dynamic ADMM regulation, and masked mapping and retraining. Second, we develop a methodology for fair and fundamental comparison of non-structured and structured pruning in terms of both storage and computation efficiency. Our results show that ADMM-NN-S consistently outperforms the prior art: 1) it achieves $348\times $ , $36\times $ , and $8\times $ overall weight pruning on LeNet-5, AlexNet, and ResNet-50, respectively, with (almost) zero accuracy loss and 2) we demonstrate the first fully binarized (for all layers) DNNs can be lossless in accuracy in many cases. These results provide a strong baseline and credibility of our study. Based on the proposed comparison framework, with the same accuracy and quantization, the results show that non-structured pruning is not competitive in terms of both storage and computation efficiency. Thus, we conclude that structured pruning has a greater potential compared to non-structured pruning. We encourage the community to focus on studying the DNN inference acceleration with structured sparsity. [ABSTRACT FROM AUTHOR]
- Published
- 2022
- Full Text
- View/download PDF
8. Zero norm based analysis model for image smoothing and reconstruction.
- Author
-
Song, Jiebo, Li, Jia, Yao, Zhengan, Ma, Kaisheng, and Bao, Chenglong
- Subjects
- *
IMAGE analysis , *IMAGE reconstruction , *SMOOTHING (Numerical analysis) , *EDGE detection (Image processing) - Abstract
The sparsity-based approaches have demonstrated promising performance in image processing. In this paper, for better preservation of the salient edge structures of images, we propose an ℓ0 + ℓ2-norm based analysis model, which requires solving a challenging non-separable ℓ0-norm related minimization problem, and we also propose an inexact augmented Lagrangian method with proven convergence to a local minimum. Extensive experiments in image smoothing, including texture removal and context smoothing, show that our method achieves better visual results over various sparsity-based models and the CNN method. Also, experiments on sparse view CT reconstruction further validate the advantage of the proposed method. [ABSTRACT FROM AUTHOR]
- Published
- 2020
- Full Text
- View/download PDF
9. Lowering Area Overheads for FeFET-Based Energy-Efficient Nonvolatile Flip-Flops.
- Author
-
Li, Xueqing, George, Sumitha, Liang, Yuhua, Ma, Kaisheng, Ni, Kai, Aziz, Ahmedullah, Gupta, Sumeet Kumar, Sampson, John, Chang, Meng-Fan, Liu, Yongpan, Yang, Huazhong, Datta, Suman, and Narayanan, Vijaykrishnan
- Subjects
- *
FIELD-effect transistors , *NONVOLATILE memory , *FERROELECTRICITY , *FLIP-flop circuits , *HAFNIUM oxide - Abstract
This brief exploits the fusion of low-power logic and nonvolatile memory inside the emerging ferroelectric FETs (FeFETs) and proposes a new nonvolatile D flip-flop (nvDFF) through the device-circuit co-design. Compared with existing FeFET-based nvDFFs with on-demand control of backup and restore (B&R), the area overhead is lowered by half, and the routing cost is reduced with embedded backup control into the supply voltage. Circuit simulations show below 5% energy-delay overhead in the normal mode and femtojoule B&R energy. This new nvDFF promises area- and energy-efficient nonvolatile computing for power-gating and energy-harvesting applications. [ABSTRACT FROM AUTHOR]
- Published
- 2018
- Full Text
- View/download PDF
10. Enabling Energy-Efficient Nonvolatile Computing With Negative Capacitance FET.
- Author
-
Li, Xueqing, Sampson, John, Khan, Asif, Ma, Kaisheng, George, Sumitha, Aziz, Ahmedullah, Gupta, Sumeet Kumar, Salahuddin, Sayeef, Chang, Meng-Fan, Datta, Suman, and Narayanan, Vijaykrishnan
- Subjects
- *
FIELD-effect transistors , *HYSTERESIS , *FERROELECTRIC materials , *ELECTRIC power failures , *ENERGY harvesting - Abstract
Negative capacitance FETs (NCFETs) have attracted significant interest due to their steep-switching capability at a low voltage and the associated benefits for implementing energy-efficient Boolean logic. While most existing works aim to avoid the ID – VG hysteresis in NCFETs, this paper exploits this hysteresis feature for logic-memory synergy and presents a custom-designed nonvolatile NCFET D flip-flop (DFF) that maintains its state during power outages. This paper also presents an NCFET fabricated for this purpose, showing <10 mV/decade steep hysteresisedges and high, up to seven orders inmagnitude, R\text {DS} ratio between the two polarization states. With a device-circuit codesign that takes advantage of the embedded nonvolatility and the high R\text {DS} ratio, the proposed DFF consumes negligible static current in backup and restore operations, and remains robust even with significant global and local ferroelectric material variations across a wide 0.3–0.8 V supply voltage range. Therefore, the proposed DFF achieves energy-efficient and low-latency backup and restore operations. Furthermore, it has an ultralow energy-delay overhead, below 2.1% in normal operations, and operates using the same voltage supply as the Boolean logic elements with which it connects. This promises energy-efficient nonvolatile computing in energy-harvesting and power-gating applications. [ABSTRACT FROM PUBLISHER]
- Published
- 2017
- Full Text
- View/download PDF
Catalog
Discovery Service for Jio Institute Digital Library
For full access to our library's resources, please sign in.