LXLixue Xia

Lixue Xia

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Lixue Xia has a total of 48 co-authors. Here are the top co-authors by number of shared publications:

  • Yu Wang

    Tsinghua University

    18 shared publications
  • Huazhong Yang

    Dept. of E.E., Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing, China

    17 shared publications
  • Tianqi Tang

    Dept. of E.E., Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing, China

    13 shared publications
  • Boxun Li

    Department of Electronic Engineering, Beijing National Research Center for Information Science and Technology (BNRist), Beijing Innovation Center for Future Chips, Tsinghua University, Beijing, China, 100084

    11 shared publications
  • Ming Cheng

    Dept. of E.E., Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing, China

    7 shared publications
  • Yu Wang

    Dept. of E.E., Tsinghua National Laboratory for Information Science and Technology (TNList), Tsinghua University, Beijing, China

    6 shared publications

Latest Publications

  • Low Bit-Width Convolutional Neural Network on RRAM

    by Yi Cai, Tianqi Tang, Lixue Xia, Boxun Li, Yu Wang, Huazhong Yang

    The emerging resistive random-access memory (RRAM) has been widely applied in accelerating the computing of deep neural networks. However, it is challenging to achieve high-precision computations based on RRAM due to the limits of the resistance level and the interfaces. Low bit-width convolutional neural networks (CNNs) provide promising solutions to introduce low bit-width RRAM devices and low bit-width interfaces in RRAM-based computing system (RCS). While open questions still remain regarding: (1) how to make matrix splitting when a single crossbar is not large enough to hold all parameters of one weight matrix; (2) how to design a pipeline to accelerate the inference based on line buffer structure; and (3) how to reduce the accuracy drop due to the parameter splitting and data quantization. In this paper, we propose an RRAM crossbar-based low bit-width CNN (LB-CNN) accelerator. We make detailed discussion on the system design, including the matrix splitting strategies to enhance the scalability, and the pipelined implementation based on line buffers to accelerate the inference. In addition, we propose a splitting and quantizing while training method to incorporate the actual hardware constraints with the training. In our experiments, low bitwidth LeNet-5 on RRAM show much better robustness than multi-bit models with device variation. The pipeline strategy achieves approximately 6.0x speedup to process each image on ResNet-18. For low-bit VGG-8 on CIFAR-10, the proposed accelerator saves 54.9% of the energy consumption and 48.3% of the area compared with the multi-bit VGG-8 structure.


  • Fault-Tolerant Training Enabled by On-Line Fault Detection for RRAM-Based Neural Computing Systems

    by Lixue Xia, Mengyun Liu, Xuefei Ning, Krishnendu Chakrabarty, Yu Wang

    An RRAM-based computing system (RCS) is an attractive hardware platform for implementing neural computing algorithms. On-line training for RCS enables hardware-based learning for a given application and reduces the additional error caused by device parameter variations. However, a high occurrence rate of hard faults due to immature fabrication processes and limited write endurance restrict the applicability of on-line training for RCS. We propose a fault-tolerant on-line training method that alternates between a fault-detection phase and a fault-tolerant training phase. In the fault-detection phase, a quiescent-voltage comparison method is utilized. In the training phase, a threshold-training method and a re-mapping scheme is proposed. Our results show that, compared to neural computing without fault tolerance, the recognition accuracy for the Cifar-10 dataset improves from 37% to 83% when using low-endurance RRAM cells, and from 63% to 76% when using RRAM cells with high endurance but a high percentage of initial faults.


  • TIME: A Training-in-Memory Architecture for RRAM-Based Deep Neural Networks

    by Ming Cheng, Lixue Xia, Zhenhua Zhu, Yi Cai, Yuan Xie, Yu Wang, Huazhong Yang

    The training of neural network (NN) is usually time-consuming and resource intensive. The emerging metal-oxide resistive random-access memory (RRAM) device has shown the potential in the computation of NN. RRAM crossbar structure and multi-bit characteristic can perform the matrix-vector product in high energy efficiency, which is the most common operation of NN. However, there exist two challenges on realizing training NN based on RRAM. Firstly, the current architectures based on RRAM only support the inference in training NN, and cannot perform the backpropagation (BP) and the weight update of training NN. Secondly, training NN requires enormous iterations to constantly update the weights for reaching the convergence. However, the weight update leads to large energy consumption because of the non-ideal factors of RRAM. In this work, we propose an architecture, TIME, which means Training-In-MEmory based on RRAM, and the peripheral circuit design to enable training NN on RRAM. TIME supports the BP and the weight update while maximizing the reuse of peripheral circuits of the inference operation on RRAM. Meanwhile, a set of optimization strategies focusing on the non-ideal factors are designed to reduce the cost of tuning RRAM. We explore the performance of both supervised learning (SL) and deep reinforcement learning (DRL) on TIME. A specific mapping method of DRL is also introduced to further improve the energy efficiency. Simulation results show that, in SL, TIME can achieve 5.3x higher energy efficiency on average compared with DaDianNao, an application-specific integrated circuits (ASIC) in CMOS technology [1]. In DRL, TIME can perform 126x on average higher than GPU in energy efficiency. If the cost of tuning RRAM can be further reduced, TIME has the potential to boost the energy efficiency by two orders of magnitudes compared with ASIC.


  • Training low bitwidth convolutional neural network on RRAM

    by Yi Cai, Tianqi Tang, Lixue Xia, Ming Cheng, Zhenhua Zhu, Yu Wang, Huazhong Yang

    Convolutional Neural Networks (CNNs) have achieved excellent performance on various artificial intelligence (AI) applications, while a higher demand on energy efficiency is required for future AI. Resistive Random-Access Memory (RRAM)-based computing system provides a promising solution to energy-efficient neural network training. However, it's difficult to support high-precision CNN in RRAM-based hardware systems. Firstly, multi-bit digital-analog interfaces will take up most energy overhead of the whole system. Secondly, it's difficult to write the RRAM to expected resistance states accurately; only low-precision numbers can be represented. To enable CNN training based on RRAM, we propose a low-bitwidth CNN training method, using low-bitwidth convolution outputs (CO), activations (A), weights (W) and gradients (G) to train CNN models based on RRAM. Furthermore, we design a system to implement the training algorithms. We explore the accuracy under different bitwidth combinations of (A, CO, W, G), and propose a practical tradeoff between accuracy and energy overhead. Our experiments demonstrate that the proposed system perform well on low-bitwidth CNN training tasks. For example, training LeNet-5 with 4-bit convolution outputs, 4-bit weights, 4-bit activations and 4-bit gradients on MNIST can still achieve 97.67% accuracy. Moreover, the proposed system can achieve 23.0X higher energy efficiency than GPU when processing the training task of LeNet-5, and 4.4X higher energy efficiency when processing the training task of ResNet-20.


  • Open Access

    Stuck-at Fault Tolerance in RRAM Computing Systems

    by Lixue Xia, Wenqin Huangfu, Tianqi Tang, Xiling Yin, Krishnendu Chakrabarty, Yuan Xie, Yu Wang, Huazhong Yang

    Emerging metal-oxide resistive switching randomaccess memory (RRAM) devices and RRAM crossbars have demonstrated their potential in boosting the speed and energyefficiency of analog matrix-vector multiplication. However, due to the immature fabrication technology, commonly occurring Stuck- At-Faults (SAFs) seriously degrade the computational accuracy of an RRAM-based Computing System (RCS). In this paper, we present a fault-tolerant framework for RCS. A Mapping Algorithm with inner fault tOlerance (MAO) is proposed to convert matrix parameters into RRAM conductances in RCS and tolerate SAFs by fully exploring the available mapping space. Two baseline redundancy schemes are proposed to ensure that RCS is effective when the percentage of faulty RRAM cells is high. To reduce the number of redundant RRAM cells when the SAFs follow a non-uniform distribution or an unknown distribution, a distribution-aware redundancy scheme and a re-configurable redundancy scheme are proposed to provide dynamic fault tolerance. Simulation results show that, the baseline redundancy schemes can improve the recognition accuracy of the MNIST dataset to almost the same as the RRAM-fault-free case, with an energy overhead of approximately 30%. When SAFs follow a non-uniform and an unknown distribution, the distributionaware and re-configurable schemes can reduce the number of redundant RRAM cells from more than 200% to less than 40% and 60%, respectively, without reducing the recognition accuracy.


  • Open Access

    TIME

    by Ming Cheng, Lixue Xia, Zhenhua Zhu, Yi Cai, Yuan Xie, Yu Wang, Huazhong Yang

    The training of neural network (NN) is usually time-consuming and resource intensive. Memristor has shown its potential in computation of NN. Especially for the metal-oxide resistive random access memory (RRAM), its crossbar structure and multi-bit characteristic can perform the matrix-vector product in high precision, which is the most common operation of NN. However, there exist two challenges on realizing the training of NN. Firstly, the current architecture can only support the inference phase of training and cannot perform the backpropagation (BP), the weights update of NN. Secondly, the training of NN requires enormous iterations and constantly updates the weights to reach the convergence, which leads to large energy consumption because of lots of write and read operations. In this work, we propose a novel architecture, TIME, and peripheral circuit designs to enable the training of NN in RRAM. TIME supports the BP and the weights update while maximizing the reuse of peripheral circuits for the inference operation on RRAM. Meanwhile, a variability-free tuning scheme and gradually-write circuits are designed to reduce the cost of tuning RRAM. We explore the performance of both SL (supervised learning) and DRL (deep reinforcement learning) in TIME, and a specific mapping method of DRL is also introduced to further improve the energy efficiency. Experimental results show that, in SL, TIME can achieve 5.3x higher energy efficiency on average compared with the most powerful application-specific integrated circuits (ASIC) in the literature. In DRL, TIME can perform averagely 126x higher than GPU in energy efficiency. If the cost of tuning RRAM can be further reduced, TIME have the potential of boosting the energy efficiency by 2 orders of magnitude compared with ASIC.


  • Top Cited Papers
    Open Access

    Fault-Tolerant Training with On-Line Fault Detection for RRAM-Based Neural Computing Systems

    by Lixue Xia, Mengyun Liu, Xuefei Ning, Krishnendu Chakrabarty, Yu Wang

    An RRAM-based computing system (RCS) is an attractive hardware platform for implementing neural computing algorithms. Online training for RCS enables hardware-based learning for a given application and reduces the additional error caused by device parameter variations. However, a high occurrence rate of hard faults due to immature fabrication processes and limited write endurance restrict the applicability of on-line training for RCS. We propose a fault-tolerant on-line training method that alternates between a fault-detection phase and a fault-tolerant training phase. In the fault-detection phase, a quiescent-voltage comparison method is utilized. In the training phase, a threshold-training method and a re-mapping scheme is proposed. Our results show that, compared to neural computing without fault tolerance, the recognition accuracy for the Cifar-10 dataset improves from 37% to 83% when using low-endurance RRAM cells, and from 63% to 76% when using RRAM cells with high endurance but a high percentage of initial faults.


  • Energy Efficient Spiking Neural Network Design with RRAM Devices

    by Yu Wang, Tianqi Tang, Boxun Li, Lixue Xia, Huazhong Yang