Yintao He

Yintao He

Auto-generated profile, based on publicly available information

Yintao He

Duke University, Durham, North Carolina, USA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications

27 April 2024
Verified
Conference Paper
Open access
SQSongyun Qu
Songyun Qu
SZShixin Zhao
Shixin Zhao
BLBing Li
Bing Li
YHYintao He
Yintao He
XCXuyi Cai
Xuyi Cai
YWYing Wang
Ying Wang

17 March 2022 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
WLWen Li
Wen Li
YWYing Wang
Ying Wang
CLCheng Liu
Cheng Liu
YHYintao He
Yintao He
LLLian Liu
Lian Liu
XLXiaowei Li
Xiaowei Li
The emerging Resistive RAM (ReRAM) technology significantly boosts the performance and the energy efficiency of the deep learning accelerators (DLAs) via the Computing-in-Memory (CiM) architecture. However, ReRAM-based DLA also suffers a high occurrence rate of memory faults. How to detect and protect against the faults in ReRAM devices poses great challenges to ReRAM-based DLA design. In this work, we propose RRAMedy, an in-situ fault detection and network remedy framework for ReRAM-based DLAs. With the proposed Adversarial Example Testing, which is a lifetime on-device and on-line fault detection technique, it achieves high detection coverage of both hard faults and soft faults at a low run-time cost. In addition, it employs an edge-cloud collaborative model retraining method to tolerate the detected faults by leveraging the inherent fault-adaptive capability of DNNs. Meanwhile, to enable in-situ model remedy when the cloud assistance is absent due to security or overhead issues, we propose to accelerate the fault-masking retraining process on edge devices with parallelized Knowledge Transfer. Our experimental results show that the proposed fault detection technique achieves high fault detection accuracy and delivers real-time testing performance. Meanwhile, the proposed retraining approach greatly alleviates the accuracy degradation problem and achieves excellent performance speedups over the baselines.

12 March 2024 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
YHYintao He
Yintao He
BLBing Li
Bing Li
YWYing Wang
Ying Wang
CLCheng Liu
Cheng Liu
HLHuawei Li
Huawei Li
XLXiaowei Li
Xiaowei Li
ReRAM-based computing-in-memory (CiM) architecture has been considered a promising solution to high-efficiency neural network accelerator, by conducting in-situ matrix multiplications and eliminating the movement of neural parameters from off-chip memory to computing units. However, we observed specific features of graph convolutional network (GCN) tasks pose design challenges to implement a high-efficiency ReRAM GCN accelerator. The ultralarge input feature data in some GCN tasks incur massive data movements, the extremely sparse adjacency matrix and input feature data involve the valid computation, and the super-large adjacency matrix that exceeds available ReRAM capacity causes frequent expensive write operations. To address the above challenges, we propose TARe, a task-adaptive CiM architecture, which consists of a hybrid in-situ computing mode to support the input feature in crossbar computing, a compact mapping scheme for efficient sparse matrix computing, and a write-free mapping to eliminate write activities in the computations with the super-large adjacency matrix. Additionally, TARe is facilitated with a task adaptive selection algorithm to generate optimized design schemes for graph neural network (GNN) tasks that have various operand sizes and data sparsity. We evaluate TARe on 11 diverse GNN tasks and compare it with different design counterparts, and the results show that achieves 168.06 $\times $ speedup and 10.95 $\times $ energy consumption reduction on average over the baseline in common GCN workloads.

23 June 2024
Verified
Conference Paper
Open access
LLLian Liu
Lian Liu
ZXZhaohui Xu
Zhaohui Xu
YHYintao He
Yintao He
YWYing Wang
Ying Wang
HLHuawei Li
Huawei Li
YHYinhe Han
Yinhe Han
Quantization is one of the most hardware-efficient ways to reduce inference costs for deep neural network (DNN) models. Nevertheless, with the continuous increase of DNN model sizes (240× in two years) and the emergence of large language models, existing static quantization methods fail to utilize the sparsity and redundancy of models sufficiently. Motivated by the pervasive dynamism in data tensors across DNN models, we propose a dynamic precision quantization algorithm to further reduce computational costs beyond statically quantized DNN models. Furthermore, we find that existing precision-flexible accelerators cannot support the DNN models with dynamic precision. To this end, we design a novel accelerator, Drift, and achieve online scheduling to efficiently support dynamic precision execution. We conduct experiments with various DNN models, including CNN-based and Transformer-based models. Evaluation results show that Drift achieves 2.85× speedup and 3.12× energy saving compared to existing precision-flexible accelerators with statically quantized models.

6 August 2021 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
YHYintao He
Yintao He
YWYing Wang
Ying Wang
HLHuawei Li
Huawei Li
XLXiaowei Li
Xiaowei Li
In-memory computing is recognized as one of the most promising architecture solution to realize energy-efficient neural network inference. Amongst many memory technology, Resistive RAM (RRAM) is a very attractive device to implement the IMC-based neural network accelerator architecture, which is particularly suitable for power-constrained IoT systems. Due to the nature of low leakage and in-situ computing, the dynamic power consumption of dot-production operations in RRAM crossbars dominates the chip power, especially when applied to lowprecision neural networks. This work investigates the correlation between the cell resistance state and the crossbar operation power, and proposes a State-Aware RRAM Accelerator (SARA) architecture for energy-efficient low-precision neural networks. With the proposed state-aware network training and mapping strategy, crossbars in the RRAM accelerator can perform in a lower-power state. Furthermore, we also leverage the proposed RRAM accelerator architecture to reduce the power consumption of high-precision network inference with both single-level or multi-level RRAM. The evaluation results show that for binary neural networks, our design saves 40.53% RRAM computing energy on average over the baseline. For high precision neural networks, the proposed method reduces 11.67% computing energy on average without any accuracy loss.