Post-training quantization reduces U-Net memory by 4x with maintained or improved segmentation accuracy, and a genetic-algorithm approach trains LUT-based binary networks for 10-15 ns FPGA inference without DSP or BRAM.
AMD zynq™UltraScale+™MPSoC ZCU104 evaluation kithttps://www.amd.com/en/ products/adaptive-socs-and-fpgas/evaluation-boards/zcu104.html
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Quantization Effects of Artificial Neural Networks for Embedded Edge-Computing Applications
Post-training quantization reduces U-Net memory by 4x with maintained or improved segmentation accuracy, and a genetic-algorithm approach trains LUT-based binary networks for 10-15 ns FPGA inference without DSP or BRAM.