Xin Li

Xin Li

Auto-generated profile, based on publicly available information

Xin Li

Huazhong University of Science and Technology

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 February 2022 in IEEE Electron Device Letters
Verified
Journal Article
Closed access
XLXin Li
Xin Li
QHQiang He
Qiang He
HTHao Tong
Hao Tong
XMXiangshui Miao
Xiangshui Miao
Phase change materials, which has been focused on the non-volatile memory field, show the possibility to carry out data storage and computing in the same physical location. However, the resistance drift behavior of phase change memory has been a huge barrier not only to traditional binary memory application for a long time, but to multi-level storage and therefore the neural network computing. Here, a bipolar programming scheme is exploited to achieve drift-reduced intermediate states and convolutional neural network (CNN) computations in Ge 2 Sb 2 Te 5 (GST) based memory cells. Experiments show that the resistance drift phenomena under bipolar programming have been reduced. Furthermore, the impact of bipolar operation on CNN for inference is investigated. This work provides effective means for implementing phase change neuromorphic processor with enhanced stability.

6 October 2023 in IEEE Transactions on Industrial Electronics
Verified
Journal Article
Closed access
ZCZiqi Chen
Ziqi Chen
XLXin Li
Xin Li
XZXun Zhu
Xun Zhu
HLHaoyuan Liu
Haoyuan Liu
HTHao Tong
Hao Tong
XMXiangshui Miao
Xiangshui Miao
Artificial neural networks (ANNs) have recently made a significant impact on the field of industrial electronics. Increasing efforts have been focused on implementing ANN using more efficient and compact architectures to achieve improved computational performance and power efficiency. The activation function of the nonlinear layers is crucial for the successful implementation of ANN, as it enables the network to possess generalization ability. In this study, we implemented the hyperbolic tangent function (tanh) using an analog circuit based on phase-change memory (PCM), which is a mature nonvolatile technology. This fully analog PCM-based tanh function is naturally compatible with analog-memory-based ANN architectures, such as the memristive neural networks (MNNs). An analog circuit has been proposed to implement the tanh by transforming the PCM I-V characteristics. The results indicate that the implemented tanh based on PCM presents a low root-mean-square error to the ideal tanh. Furthermore, the performance of the implemented tanh was verified in a classical deep neural network (DNN) with the handwritten digit recognition task. The implemented tanh with realistic characteristics presents a training accuracy (>90%) and good precision for repeatability inference in DNN. This work provides effective means for implementing activation functions in ANN based on analog memory.

29 June 2023 in Materials Today Electronics
Verified
Journal Article
Open access
ZYZhe Yang
Zhe Yang
ZYZiyan Yang
Ziyan Yang
LLLong Liu
Long Liu
XLXin Li
Xin Li
JLJunze Li
Junze Li
XMXiangshui Miao
Xiangshui Miao

17 July 2023 in Applied Physics Letters
Verified
Journal Article
Open access
PDF file
XLXin Li
Xin Li
RZRuizhe Zhao
Ruizhe Zhao
HTHao Tong
Hao Tong
XMXiangshui Miao
Xiangshui Miao
Phase change memory (PCM) is one of the most mature technologies for non-von Neumann computing. However, abrupt amorphization becomes a barrier for training artificial neural networks, due to limitations of the inherent operational mechanism of phase change materials. The devices can achieve a gradual conductance change in the crystallization process, while the conductance change for amorphization process is much more abrupt. This work presents a possible explanation for the RESET abrupt change issue in T-shaped devices, based on the analysis of the volume and connectivity of the amorphous and crystalline regions. Using this model, a nanoribbon device for analog PCM targeting neural network applications is designed, fabricated, and characterized. The designed device can realize a gradual RESET without changing the amplitude and width of RESET pulses. Using a nanoribbon device as a single synapse in the designed array reduces the number of SET operations needed to achieve the same accuracy in convolutional neural network simulation by 75%, which implies a significant reduction in power and time consumption. This work provides an effective way to implement gradual RESET for PCM devices.

23 May 2025 in IEEE Transactions on Electron Devices
Verified
Journal Article
Closed access
YCYilin Chen
Yilin Chen
XLXin Li
Xin Li
ZCZiqi Chen
Ziqi Chen
HTHao Tong
Hao Tong
XMXiangshui Miao
Xiangshui Miao
Activation function (AF) is a pivotal yet resource-intensive component in artificial neural network (ANN) hardware implementations. Digital AFs offer high accuracy but require data conversions incurring hardware overhead in analog-memory-based ANNs, making analog AFs more suitable. However, existing analog approaches leveraging device physics for direct AF implementations face significant challenges in integration and scalability, underscoring the urgent need to explore alternative memory devices for more compact and reliable AF implementations. This work proposes a low-power, high speed AF constructed directly using the inherent Sinh current-to-voltage (I–V) characteristics of phase-change memory (PCM) devices. We designed an analog circuit to implement the inverse hyperbolic sine (Arsinh) function and achieved the target AF through the rapid read operation of a single PCM cell. Experimental results demonstrate that the proposed Arsinh function exhibits a low root-mean-square error (RMSE) compared to the ideal Arsinh function, with a highly efficient operational power consumption ranging from ~0.17 nW to 3.5 μW. Additionally, the delay time (~70 ns) is comparable to the speed of matrix–vector multiplications (MVMs) in mature System-on-Chip (SoC) implementations. Furthermore, the AF achieved a classification accuracy of 94.81% on the MNIST dataset and 91.01% on the Fashion MNIST dataset, confirming its effectiveness and applicability in practical scenarios.

1 February 2023 in arXiv
Verified
Preprint
Open access
PDF file
RZRunhua Zhang
Runhua Zhang
HJHongxu Jiang
Hongxu Jiang
TFTian Fangzheng
Tian Fangzheng
Jinkun GengJinkun Geng
Jinkun Geng
LXLi Xiaobin
Li Xiaobin
WHWang Haojie
Wang Haojie
Edge computing has been emerging as a popular scenario for model inference. However, the inference performance on edge devices (e.g., Multi-Core DSP, FGPA, etc.) suffers from inefficiency due to the lack of highly optimized inference frameworks. Previous model inference frameworks are mainly developed in an operator-centric way, which provides insufficient acceleration to edge-based inference. Besides, the operator-centric framework incurs significant costs for continuous development and maintenance. In this paper, we propose Xenos, which can automatically conduct dataflow-centric optimization of the computation graph and accelerate inference in two dimensions. Vertically, Xenos develops operator linking technique to improve data locality by restructuring the inter-operator dataflow. Horizontally, Xenos develops DSP-aware operator split technique to enable higher parallelism across multiple DSP units. Our evaluation proves the effectiveness of vertical and horizontal dataflow optimization, which reduce the inference time by 21.2\%--84.9\% and 17.9\%--96.2\% , respectively. Besides, Xenos also outperforms the widely-used TVM by 3.22$\times$--17.92$\times$. Moreover, we extend Xenos to a distributed solution, which we call d-Xenos. d-Xenos employs multiple edge devices to jointly conduct the inference task and achieves a speedup of 3.68x--3.78x compared with the single device.