47 citations · 108 across the 16 of their papers we have counts for
12 papers
COAC: Cross-layer Optimization of Accelerator Configurability for Efficient CNN Processing
Steven Colleman, Man Shi, Marian Verhelst
To achieve high accuracy, convolutional neural networks (CNNs) are increasingly growing in complexity and diversity in layer types and topologies. This makes it very challenging to…
Optimising GPGPU Execution Through Runtime Micro-Architecture Parameter Analysis
Giuseppe M. Sarda, Nimish Shah, Debjyoti Bhattacharjee +2
GPGPU execution analysis has always been tied to closed-source, proprietary benchmarking tools that provide high-level, non-exhaustive, and/or statistical information, preventing a…
CMDS: Cross-layer Dataflow Optimization for DNN Accelerators Exploiting Multi-bank Memories
Man Shi, Steven Colleman, Charlotte VanDeMieroop +4
Deep neural networks (DNN) use a wide range of network topologies to achieve high accuracy within diverse applications. This model diversity makes it impossible to identify a singl…
Optimizing Layer-Fused Scheduling of Transformer Networks on Multi-accelerator Platforms
Steven Colleman, Arne Symons, Victor J. B. Jung +1
The impact of transformer networks is booming, yet, they come with significant computational complexity. It is therefore essential to understand how to optimally map and execute th…
HTVM: Efficient Neural Network Deployment On Heterogeneous TinyML Platforms
Josse Van Delm, Maarten Vandersteegen, Alessio Burrello +5
Optimal deployment of deep neural networks (DNNs) on state-of-the-art Systems-on-Chips (SoCs) is crucial for tiny machine learning (TinyML) at the edge. The complexity of these SoC…
ACCO: Automated Causal CNN Scheduling Optimizer for Real-Time Edge Accelerators
Jun Yin, Linyan Mei, Andre Guntoro +1
Spatio-Temporal Convolutional Neural Networks (ST-CNN) allow extending CNN capabilities from image processing to consecutive temporal-pattern recognition. Generally, state-of-the-a…