Hao Zheng
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
Research Subjects
Hao Zheng has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Ahmed Louri
George Washington University, Washington, DC, USA
8 shared publications - Ke Wang
University of North Carolina at Charlotte, Charlotte, NC, USA
7 shared publications - Jiajun Li
School of Astronautics, Beihang University, Beijing, China
4 shared publications - Lingxiang Yin
University of Central Florida, Orlando, USA
4 shared publications Ahmed Louri
Department of Electrical and Computer Engineering, George Washington University, Washington, DC, USA
3 shared publications- Amir Ghazizadeh Ahsaei
University of Central Florida, Orlando, FL, USA
3 shared publications
Latest Publications
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
PanGaea: A Unified Memory-Efficient Accelerator for Pangenome Chaining and Alignment
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-7
Open AccessPanGaea: A Unified Memory-Efficient Accelerator for Pangenome Chaining and Alignment
- Open Access
- 17 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 1535-1548
Rethinking Tiling and Dataflow for SpMM Acceleration: A Graph Transformation Framework
- 17 October 2025
- Published by Association for Computing Machinery (ACM)
- p. 1535-1548
Open AccessRethinking Tiling and Dataflow for SpMM Acceleration: A Graph Transformation Framework
- Open Access
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1240-1253
DiTile-DGNN: An Efficient Accelerator for Distributed Dynamic Graph Neural Network Inference
- 20 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 1240-1253
Open AccessDiTile-DGNN: An Efficient Accelerator for Distributed Dynamic Graph Neural Network Inference
- Open Access
- 4 February 2025
- Published by MDPI AG in Mathematics
- Vol. 13 (3) , 516
Effectiveness of PEER Intervention on Older Adults’ Physical Activity Time Series Using Smoothing Spline ANOVA
Falls are a major cause of injury among older adults. The Physio-fEedback Exercise pRogram (PEER) combines physio-feedback, cognitive reframing, and guided exercises to reduce fall risk. However, its impact on physical activity (PA) over time is underexplored. Functional time-series analysis offers insight into behavior patterns and sustainability. This preliminary study assessed PEER’s effectiveness in improving PA levels immediately and over time. A total of 64 community-dwelling older adults were cluster-randomized into PEER () or control groups (). Participants wore Fitbit trackers, generating time-series data on activity. The PEER group completed an 8-week program, while the control group received CDC fall prevention pamphlets. PA data were analyzed using smoothing spline analysis of variance (SSANOVA), chosen for its flexibility in modeling complex, non-linear relationships in time-series data and its ability to handle skewed distributions and repeated measures. Unlike traditional parametric models, SSANOVA decomposes temporal trends into interpretable components, capturing both smooth trends and abrupt changes, such as those occurring on group workout days. This capability ensures robust and nuanced analysis of intervention effects. Results showed PEER participants significantly increased evenly and had very active minutes and reduced sedentary behavior during the intervention. No significant effect was found for light active minutes. Specifically, during the intervention period, PEER participants engaged in an average of fewer sedentary minutes per day, additional fairly active minutes per day, and additional very active minutes per day compared to the control group. While the reduction in sedentary minutes and increase in fairly active minutes were not statistically significant, the increase in very active minutes was significant. However, our functional time-series analysis revealed these improvements diminished over the 15-week follow-up, indicating challenges in maintaining PA. In conclusion, PEER boosts PA and reduces sedentary behavior short-term, but strategies are needed to sustain these benefits. In conclusion, PEER boosts PA and reduces sedentary behavior short-term, but strategies are needed to sustain these benefits. Public health policies should emphasize technology-driven fall risk assessments, community-based prevention programs, and initiatives that promote physical activity, home safety, and chronic condition management.- 4 February 2025
- Published by MDPI AG in Mathematics
- Vol. 13 (3) , 516
Open AccessEffectiveness of PEER Intervention on Older Adults’ Physical Activity Time Series Using Smoothing Spline ANOVA
Falls are a major cause of injury among older adults. The Physio-fEedback Exercise pRogram (PEER) combines physio-feedback, cognitive reframing, and guided exercises to reduce fall risk. However, its impact on physical activity (PA) over time is underexplored. Functional time-series analysis offers insight into behavior patterns and sustainability. This preliminary study assessed PEER’s effectiveness in improving PA levels immediately and over time. A total of 64 community-dwelling older adults were cluster-randomized into PEER () or control groups (). Participants wore Fitbit trackers, generating time-series data on activity. The PEER group completed an 8-week program, while the control group received CDC fall prevention pamphlets. PA data were analyzed using smoothing spline analysis of variance (SSANOVA), chosen for its flexibility in modeling complex, non-linear relationships in time-series data and its ability to handle skewed distributions and repeated measures. Unlike traditional parametric models, SSANOVA decomposes temporal trends into interpretable components, capturing both smooth trends and abrupt changes, such as those occurring on group workout days. This capability ensures robust and nuanced analysis of intervention effects. Results showed PEER participants significantly increased evenly and had very active minutes and reduced sedentary behavior during the intervention. No significant effect was found for light active minutes. Specifically, during the intervention period, PEER participants engaged in an average of fewer sedentary minutes per day, additional fairly active minutes per day, and additional very active minutes per day compared to the control group. While the reduction in sedentary minutes and increase in fairly active minutes were not statistically significant, the increase in very active minutes was significant. However, our functional time-series analysis revealed these improvements diminished over the 15-week follow-up, indicating challenges in maintaining PA. In conclusion, PEER boosts PA and reduces sedentary behavior short-term, but strategies are needed to sustain these benefits. In conclusion, PEER boosts PA and reduces sedentary behavior short-term, but strategies are needed to sustain these benefits. Public health policies should emphasize technology-driven fall risk assessments, community-based prevention programs, and initiatives that promote physical activity, home safety, and chronic condition management. - Open Access
- 27 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-9
CircuitSeer: RTL Post-PnR Delay Prediction via Coupling Functional and Structural Representation
- 27 October 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-9
Open AccessCircuitSeer: RTL Post-PnR Delay Prediction via Coupling Functional and Structural Representation
- Open Access
- 23 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
EGMA: Enhancing Data Reuse and Workload Balancing in Message Passing GNN Acceleration via Gram Matrix Optimization
Graph Neural Networks (GNNs) have been widely used to handle intricate graph-related problems, in which complex vertex and edge operations are performed in the form of message passing between vertices. Such complex GNN operations are highly dependent on the graph structure and can no longer be characterized as sparse-dense or general matrix multiplications. Consequently, current matrix-based data reuse and workload balancing optimizations have limited applicability to Message Passing-based GNN acceleration. In this paper, we leverage the mathematical insights from Gram Matrix to simultaneously exploit data reuse and workload balancing opportunities for message passing-based GNN accelerations. Upon this insight, we further propose a novel accelerator, named EGMA, that can efficiently facilitate a wide range of GNN models with improved data reuse and workload balance. Consequently, EGMA can achieve performance speedup by 1.57×, 1.72×, and 1.43× and energy reduction by 38.19%, 34.02%, and 24.54% on average compared to Betty, FlowGNN, and ReGNN, respectively.
- 23 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessEGMA: Enhancing Data Reuse and Workload Balancing in Message Passing GNN Acceleration via Gram Matrix Optimization
Graph Neural Networks (GNNs) have been widely used to handle intricate graph-related problems, in which complex vertex and edge operations are performed in the form of message passing between vertices. Such complex GNN operations are highly dependent on the graph structure and can no longer be characterized as sparse-dense or general matrix multiplications. Consequently, current matrix-based data reuse and workload balancing optimizations have limited applicability to Message Passing-based GNN acceleration. In this paper, we leverage the mathematical insights from Gram Matrix to simultaneously exploit data reuse and workload balancing opportunities for message passing-based GNN accelerations. Upon this insight, we further propose a novel accelerator, named EGMA, that can efficiently facilitate a wide range of GNN models with improved data reuse and workload balance. Consequently, EGMA can achieve performance speedup by 1.57×, 1.72×, and 1.43× and energy reduction by 38.19%, 34.02%, and 24.54% on average compared to Betty, FlowGNN, and ReGNN, respectively.
- Open Access
- 23 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
VITA: ViT Acceleration for Efficient 3D Human Mesh Recovery via Hardware-Algorithm Co-Design
Vision Transformers (ViTs) have emerged as a promising solution to enable efficient 3D Human Mesh Recovery (HMR) in augmented and virtual reality (AR/VR) applications. Despite many advancements in algorithm design, it remains a challenge to efficiently accelerate ViT-based HMR due to high computational complexity, substantial memory footprint, and compromised data locality. In this paper, we propose VITA, a hardware and algorithm co-design framework for ViT-based HMR with improved performance and energy efficiency. Specifically, on the algorithm side, we propose an average pooling model to replace conventional multi-head attention, which is further optimized with improved data locality. On the hardware side, we propose an accelerator architecture that can efficiently support various dataflows and computations demanded by pooling, normalization, and convolution operations. We evaluate the proposed VITA, and the evaluation result shows that the proposed VITA design can achieve 5.05× and 69.12× speedups on average over the state-of-the-art GPUs and CPUs on HMR tasks.
- 23 June 2024
- Published by Association for Computing Machinery (ACM)
- p. 1-6
Open AccessVITA: ViT Acceleration for Efficient 3D Human Mesh Recovery via Hardware-Algorithm Co-Design
Vision Transformers (ViTs) have emerged as a promising solution to enable efficient 3D Human Mesh Recovery (HMR) in augmented and virtual reality (AR/VR) applications. Despite many advancements in algorithm design, it remains a challenge to efficiently accelerate ViT-based HMR due to high computational complexity, substantial memory footprint, and compromised data locality. In this paper, we propose VITA, a hardware and algorithm co-design framework for ViT-based HMR with improved performance and energy efficiency. Specifically, on the algorithm side, we propose an average pooling model to replace conventional multi-head attention, which is further optimized with improved data locality. On the hardware side, we propose an accelerator architecture that can efficiently support various dataflows and computations demanded by pooling, normalization, and convolution operations. We evaluate the proposed VITA, and the evaluation result shows that the proposed VITA design can achieve 5.05× and 69.12× speedups on average over the state-of-the-art GPUs and CPUs on HMR tasks.
- 8 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 35 (2) , 349-361
Versa-DNN: A Versatile Architecture Enabling High-Performance and Energy-Efficient Multi-DNN Acceleration
Emerging applications utilize numerous Deep Neural Networks (DNNs) to address multiple tasks simultaneously. As these applications continue to expand, there is a growing need for off-chip memory access optimization and innovative architectures that can adapt to diverse computation, memory, and communication requirements of various DNN models. To address these challenges, we propose Versa-DNN, a versatile DNN accelerator that can provide efficient computation, memory, and communication support for the simultaneous execution of multiple DNNs. Versa-DNN features three unique designs: a flexible off-chip memory access optimization strategy, adaptable communication fabrics, and a communication and computational aware scheduling algorithm. The proposed off-chip memory optimization strategy can improve performance and energy efficiency by increasing hardware utilization, eliminating excess data duplication, and reducing off-chip memory accesses. The adaptable communication fabrics consist of distributed buffers, processing elements, and a flexible Network-on-Chip (NoC), which can dynamically morph and fission to support distinct communication and computation needs for simultaneously running DNN models. Furthermore, the proposed scheduling policy manages the simultaneous execution of multiple DNN models with improved performance and energy efficiency. Simulation results using several DNN models, show that the proposed Versa-DNN architecture achieves 41%, 238%, 392% throughput speedup and 30%, 59%, 63% energy reduction on average for different workloads when compared to state-of-the-art accelerators such as Planaria, Herald, and AI-MT, respectively.
- 8 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 35 (2) , 349-361
Versa-DNN: A Versatile Architecture Enabling High-Performance and Energy-Efficient Multi-DNN Acceleration
Emerging applications utilize numerous Deep Neural Networks (DNNs) to address multiple tasks simultaneously. As these applications continue to expand, there is a growing need for off-chip memory access optimization and innovative architectures that can adapt to diverse computation, memory, and communication requirements of various DNN models. To address these challenges, we propose Versa-DNN, a versatile DNN accelerator that can provide efficient computation, memory, and communication support for the simultaneous execution of multiple DNNs. Versa-DNN features three unique designs: a flexible off-chip memory access optimization strategy, adaptable communication fabrics, and a communication and computational aware scheduling algorithm. The proposed off-chip memory optimization strategy can improve performance and energy efficiency by increasing hardware utilization, eliminating excess data duplication, and reducing off-chip memory accesses. The adaptable communication fabrics consist of distributed buffers, processing elements, and a flexible Network-on-Chip (NoC), which can dynamically morph and fission to support distinct communication and computation needs for simultaneously running DNN models. Furthermore, the proposed scheduling policy manages the simultaneous execution of multiple DNN models with improved performance and energy efficiency. Simulation results using several DNN models, show that the proposed Versa-DNN architecture achieves 41%, 238%, 392% throughput speedup and 30%, 59%, 63% energy reduction on average for different workloads when compared to state-of-the-art accelerators such as Planaria, Herald, and AI-MT, respectively.
- Open Access
- 19 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 97-104
CryptoTrain: Fast Secure Training on Encrypted Dataset
Secure training, while protecting the confidentiality of both data and model weights, typically incurs significant training overhead. Traditional Fully Homomorphic Encryption (FHE)-based non-inter-active training models are heavily burdened by computationally demanding bootstrapping. To develop an efficient secure training system, we established a foundational framework, CryptoTrain-B, utilizing a hybrid cryptographic protocol that merges FHE with Oblivious Transfer (OT) for handling linear and non-linear operations, respectively. This integration eliminates the need for costly bootstrapping. Although CryptoTrain-B sets a new baseline in performance, reducing its training overhead remains essential. We found that ciphertext-ciphertext multiplication (CCMul) is a critical bottleneck in operations involving encrypted inputs and models. Our solution, the CCMul-Precompute technique, involves precomputing CCMul offline and resorting to the less resource-intensive ciphertext-plaintext multiplication (CPMul) during private training. Furthermore, conventional polynomial convolution in FHE systems tends to encode irrelevant and redundant values into polynomial slots, necessitating additional polynomials and ciphertexts for input representation and leading to extra multiplications. Addressing this, we introduce correlated polynomial convolution, which encodes only related input values into polynomials, thus drastically reducing the number of computations and overheads. By integrating CCMul-Precompute and correlated polynomial convolution into CryptoTrain-B, we facilitate a rapid and efficient secure training framework, CryptoTrain. Extensive experiments demonstrate that CryptoTrain achieves a ~5.3× training time reduction compared to prior methods.
- 19 November 2023
- Published by Association for Computing Machinery (ACM)
- p. 97-104
Open AccessCryptoTrain: Fast Secure Training on Encrypted Dataset
Secure training, while protecting the confidentiality of both data and model weights, typically incurs significant training overhead. Traditional Fully Homomorphic Encryption (FHE)-based non-inter-active training models are heavily burdened by computationally demanding bootstrapping. To develop an efficient secure training system, we established a foundational framework, CryptoTrain-B, utilizing a hybrid cryptographic protocol that merges FHE with Oblivious Transfer (OT) for handling linear and non-linear operations, respectively. This integration eliminates the need for costly bootstrapping. Although CryptoTrain-B sets a new baseline in performance, reducing its training overhead remains essential. We found that ciphertext-ciphertext multiplication (CCMul) is a critical bottleneck in operations involving encrypted inputs and models. Our solution, the CCMul-Precompute technique, involves precomputing CCMul offline and resorting to the less resource-intensive ciphertext-plaintext multiplication (CPMul) during private training. Furthermore, conventional polynomial convolution in FHE systems tends to encode irrelevant and redundant values into polynomial slots, necessitating additional polynomials and ciphertexts for input representation and leading to extra multiplications. Addressing this, we introduce correlated polynomial convolution, which encodes only related input values into polynomials, thus drastically reducing the number of computations and overheads. By integrating CCMul-Precompute and correlated polynomial convolution into CryptoTrain-B, we facilitate a rapid and efficient secure training framework, CryptoTrain. Extensive experiments demonstrate that CryptoTrain achieves a ~5.3× training time reduction compared to prior methods.
- 11 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Sustainable Computing
- Vol. 9 (2) , 115-127
Morph-GCNX: A Universal Architecture for High-Performance and Energy-Efficient Graph Convolutional Network Acceleration
While current Graph Convolutional Networks (GCNs) accelerators have achieved notable success in a wide range of application domains, these GCN accelerators can not support various intra- and inter- GCN dataflows or adapt to diverse GCN applications. In this paper, we propose Morph-GCNX, a flexible GCN accelerator architecture for high-performance and energy-efficient GCN execution. The proposed design consists of a flexible Processing Element (PE) array that can be partitioned at runtime and adapt to the computational needs of different layers within a GCN or multiple concurrent GCNs. The proposed Morph-GCNX also consists of a morphable interconnection design to support a wide range of GCN dataflows with various parallelization and data reuse strategies for GCN execution. We also propose a hardware-application co-exploration technique that explores the GCN and hardware design spaces to identify the best PE partition, workload allocation, dataflow, and interconnection configurations, with the goal of improving overall performance and energy. Simulation results show that the proposed Morph-GCNX architecture achieves 18.8×, 2.9×, 1.9×, 1.8×, and 2.5× better performance, reduces DRAM accesses by a factor of 10.8×, 3.7×, 2.2×, 2.5×, and 1.3×, and improves energy consumption by 13.2×, 5.6×, 2.1×, 2.5×, and 1.3×, as compared to prior designs including HyGCN, AWB-GCN, LW-GCN, GCoD, and GCNAX, respectively.
- 11 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Sustainable Computing
- Vol. 9 (2) , 115-127
Morph-GCNX: A Universal Architecture for High-Performance and Energy-Efficient Graph Convolutional Network Acceleration
While current Graph Convolutional Networks (GCNs) accelerators have achieved notable success in a wide range of application domains, these GCN accelerators can not support various intra- and inter- GCN dataflows or adapt to diverse GCN applications. In this paper, we propose Morph-GCNX, a flexible GCN accelerator architecture for high-performance and energy-efficient GCN execution. The proposed design consists of a flexible Processing Element (PE) array that can be partitioned at runtime and adapt to the computational needs of different layers within a GCN or multiple concurrent GCNs. The proposed Morph-GCNX also consists of a morphable interconnection design to support a wide range of GCN dataflows with various parallelization and data reuse strategies for GCN execution. We also propose a hardware-application co-exploration technique that explores the GCN and hardware design spaces to identify the best PE partition, workload allocation, dataflow, and interconnection configurations, with the goal of improving overall performance and energy. Simulation results show that the proposed Morph-GCNX architecture achieves 18.8×, 2.9×, 1.9×, 1.8×, and 2.5× better performance, reduces DRAM accesses by a factor of 10.8×, 3.7×, 2.2×, 2.5×, and 1.3×, and improves energy consumption by 13.2×, 5.6×, 2.1×, 2.5×, and 1.3×, as compared to prior designs including HyGCN, AWB-GCN, LW-GCN, GCoD, and GCNAX, respectively.