An Open-Source ML-Based Full-Stack Optimization Framework for Machine Learning Accelerators
Parameterizable machine learning (ML) accelerators are the product of recent
breakthroughs in ML. To fully enable their design space exploration (DSE), we propose a …
breakthroughs in ML. To fully enable their design space exploration (DSE), we propose a …
Physically accurate learning-based performance prediction of hardware-accelerated ml algorithms
Parameterizable ML accelerators are the product of recent breakthroughs in machine
learning (ML). To fully enable the design space exploration, we propose a physical-design …
learning (ML). To fully enable the design space exploration, we propose a physical-design …
ARCO: Adaptive Multi-Agent Reinforcement Learning-Based Hardware/Software Co-Optimization Compiler for Improved Performance in DNN Accelerator Design
This paper presents ARCO, an adaptive Multi-Agent Reinforcement Learning (MARL)-based
co-optimizing compilation framework designed to enhance the efficiency of mapping …
co-optimizing compilation framework designed to enhance the efficiency of mapping …
Reusing GEMM hardware for efficient execution of depthwise separable convolution on ASIC-based DNN accelerators
SD Manasi, S Banerjee, A Davare, AA Sorokin… - Proceedings of the 28th …, 2023 - dl.acm.org
Deep learning (DL) accelerators are optimized for standard convolution. However,
lightweight convolutional neural networks (CNNs) use depthwise convolution (DwC) in key …
lightweight convolutional neural networks (CNNs) use depthwise convolution (DwC) in key …
Performance Analysis of DNN Inference/Training with Convolution and non-Convolution Operations
Today's performance analysis frameworks for deep learning accelerators suffer from two
significant limitations. First, although modern convolutional neural network (CNNs) consist of …
significant limitations. First, although modern convolutional neural network (CNNs) consist of …
Performance analysis of CNN inference/training with convolution and non-convolution operations on ASIC accelerators
Today's performance analysis frameworks for deep learning accelerators suffer from two
significant limitations. First, although modern convolutional neural networks (CNNs) consist …
significant limitations. First, although modern convolutional neural networks (CNNs) consist …
DNN Model Theft Through Trojan Side-Channel on Edge FPGA Accelerator
S Chandrasekar, SK Lam, S Thambipillai - International Symposium on …, 2023 - Springer
In this paper, we present a novel hardware trojan assisted side-channel attack to reverse
engineer DNN architectures on edge FPGA accelerators. In particular, our attack targets the …
engineer DNN architectures on edge FPGA accelerators. In particular, our attack targets the …
Optimization of the Versatile Tensor Accelerator (VTA) Load Module in a Time-Triggered Memory Access
AM Ezekiel, D Onwuchekwa… - 2023 26th Euromicro …, 2023 - ieeexplore.ieee.org
Embedded systems powered by artificial intelligence (AI) are widely employed in diverse
domains. However, the lack of inherent predictability in existing AI accelerators poses …
domains. However, the lack of inherent predictability in existing AI accelerators poses …
Exploration for Efficient Depthwise Separable Convolution Networks Deployment on FPGA
Z Huang, A Qie, C Zhang, J Yang… - 2024 IEEE 6th …, 2024 - ieeexplore.ieee.org
Depthwise Separable Convolution (DSC) has become the key structure in lightweight
convolutional neural networks. However, the tight connection between network structure and …
convolutional neural networks. However, the tight connection between network structure and …
Software-driven Design for Domain-specific Compute
DA Kirkpatrick - Proceedings of the 2023 International Symposium on …, 2023 - dl.acm.org
The end of Dennard scaling has created a focus on advancing domain-specific computing;
we are seeing a renaissance of accelerating compute problems through specialization, with …
we are seeing a renaissance of accelerating compute problems through specialization, with …