A5103205936- Parallel Computing and Optimization Techniques
- Embedded Systems Design Techniques
- Advanced Memory and Neural Computing
- Ferroelectric and Negative Capacitance Devices
- Biosensors and Analytical Detection
- Interconnection Networks and Systems
- Caching and Content Delivery
- Advanced biosensing and bioanalysis techniques
- Green IT and Sustainability
- Cloud Computing and Resource Management
- thermodynamics and calorimetric analyses
- Real-Time Systems Scheduling
- Environmental Impact and Sustainability
- Advanced Graph Neural Networks
- Real-time simulation and control systems
- Modular Robots and Swarm Intelligence
- Fuel Cells and Related Materials
- Optical Imaging and Spectroscopy Techniques
- Cancer Genomics and Diagnostics
- Genetic factors in colorectal cancer
- Machine Learning in Materials Science
- CCD and CMOS Imaging Sensors
- Brain Tumor Detection and Classification
- CRISPR and Genetic Engineering
- Genomics and Rare Diseases
Brown University
2024-2025
Changchun Normal University
2023-2025
University of Pittsburgh
2024
Dalian University of Technology
2024
John Brown University
2024
Jilin Business and Technology College
2023
Jilin Province Science and Technology Department
2023
Three-way light guides containing one or more strands of 25-micron 80-micron diameter optical fibers in each channel have been constructed and used to measure the NADH fluorescence UV reflectance from mitochondrial suspensions, perfused, hemoglobin-free rat liver, perfused beating interventricular septum rabbit. The changes measured with these so-called micro-light guides, which channels several less than 100 micron, are comparable magnitude those using much larger conventional guides....
With the increase in computation intensity of chip, mismatch between layer shapes and available resource significantly limits utilization chip. Driven by this observation, prior works discuss spatial accelerators or dataflow architecture to maximize throughput. However, using could potentially execution latency. In work, we first systematically investigate two models: (1) sequentially (temporally) launch one monolithic accelerator, (2) spatially multiple accelerators. From observations, find...
While vision transformers (ViTs) have shown consistent progress in computer vision, deploying them for real-time decision-making scenarios (<1 ms) is challenging. Current computing platforms like CPUs, GPUs, or FPGA-based solutions struggle to meet this deterministic low-latency requirement, even with quantized ViT models. Some approaches use pruning sparsity reduce the model size and latency, but often results accuracy loss. To address aforementioned constraints, work, we propose EQ-ViT, an...
Fast-evolving artificial intelligence (AI) algorithms such as large language models have been driving the ever-increasing computing demands in today's data centers. Heterogeneous with domain-specific architectures (DSAs) brings many opportunities when scaling up and out system. In particular, heterogeneous chiplet architecture is favored to keep system well reduce design complexity cost stemming from traditional monolithic chip design. However, how interconnect resources orchestrate chiplets...
Fast-evolving artificial intelligence (AI) algorithms such as large language models have been driving the ever-increasing computing demands in today's data centers. Heterogeneous with domain-specific architectures (DSAs) brings many opportunities when scaling up and out system. In particular, heterogeneous chiplet architecture is favored to keep system well reduce design complexity cost stemming from traditional monolithic chip design. However, how interconnect resources orchestrate chiplets...
methods for screening hazardous chemicals are necessary sound management. Persistent, bioaccumulative, mobile, and toxic (PBMT) persist in the environment have high mobility aquatic environments, posing risks to human ecological health. However, lack of experimental data vast number hinders identification PBMT chemicals. Through an extensive search measured chemical data, as well persistent, (PBT) inventories, this study constructed comprehensive sets on To address limited volume set, a...
Embodied carbon has been widely reported as a significant component in the full system lifecycle of various computing systems green house gas emissions. Many efforts have undertaken to quantify elements that comprise this embodied carbon, from tools evaluate semiconductor manufacturing those can different commercial and academic sources. However, these cannot easily reproduce results by server vendors' product reports accuracy vary substantially due assumptions. Furthermore, attempts...
Recently, the increasing need for computing resources has led to prosperity of data centers, which poses challenges environmental impacts and calls improvements in center provisioning strategies. In this work, we show a comprehensive analysis based on profiling variety deep-learning inference applications different generations GPU servers. Our reveals several critical factors can largely affect design space strategies including hardware embodied cost estimation, application-specific...
Dense matrix multiply (MM) serves as one of the most heavily used kernels in deep learning applications. To cope with high computation demands these applications, heterogeneous architectures featuring both FPGA and dedicated ASIC accelerators have emerged promising platforms. For example, AMD/Xilinx Versal ACAP architecture combines general-purpose CPU cores programmable logic AI Engine processors optimized for AI/ML. An array 400 executing at 1 GHz can provide up to 6.4 TFLOPS performance...
Graph Convolutional Networks (GCNs) are widely used in graph-based applications, such as social networks and recommendation systems. Nevertheless, large-scale graphs or deep aggregation layers full-batch GCNs consume significant GPU memory, causing out of memory (OOM) errors on mainstream GPUs (e.g., 29GB consumption the Ogbnproducts graph with 5 layers). The subgraph sampling methods reduce to achieve lightweight by partitioning into multiple subgraphs sequentially training each subgraph....
Cognitive Diagnosis (CD) aims to evaluate students' cognitive states based on their interaction data, enabling downstream applications such as exercise recommendation and personalized learning guidance. However, existing methods often struggle with accuracy drops in cross-domain diagnosis (CDCD), a practical yet challenging task. While some efforts have explored exercise-aspect CDCD, crosssubject scenarios, they fail address the broader dual-aspect nature of encompassing both student-...