arXiv · cs.AI · Jinsong Zhang, Minghe Li, Jiayi Tian, Jinming Lu, Zheng Zhang · 10 小时前模型评分37
超越 MACs：面向张量化神经网络的部署约束设计空间探索
Beyond MACs: Deployment-Constrained Design Space Exploration for Tensorized Neural Networks
导读
研究提出一种面向固定秩张量化神经网络（TNN）的部署约束、延迟引导设计空间探索框架，在固定处理单元（PE）预算下联合评估收缩路径、三种 PE 阵列划分与数据流组合。
分类置信度
中 · 待人工复核
- 只有单一媒体来源，需要人工复核。
来源级别：媒体与个人
事件类型：研究发布
证据摘录
- “We present a deployment-constrained, latency-guided design space exploration (DSE) framework for fixed-rank tensorized neural networks (TNNs) under a fixed processing-element (PE) budget.”
- “On ResNet-18 and ViT-Ti/4, 25%-50% of the layers select a path other than the minimum-MAC one, the best dataflow varies across layers and models, and on a representative ViT layer the latency-selected candidate reduces simulated latency by 23.5% relative to the minimum-MAC candidate.”
- “Deployed on a Xilinx VU9P FPGA accelerator that supports both inference and training, the complete TT+INT8 system achieves 3.28x-4.00x inference and 3.42x-3.85x training speedups over dense implementations with the same PE budget.”
来源：arXiv · cs.AI · arxiv.org