DEX: Digit-Level Early Exit for Energy-Efficient MSDF Neural Network Inference
U-Net inference for brain-tumor segmentation requires billions of multiply-accumulate operations, motivating hardware that can reduce computation dynamically rather than relying only on fixed precision or static model compression.
Key points
- Most-significant-digit-first (MSDF) arithmetic exposes the leading digits of a result during computation, enabling output-dependent decisions before the full value is generated.
- This paper presents an MSDF accelerator for quantized U-Net segmentation with a two-stage grouped processing element supporting signed INT8 operands and in-stream bias accumulation.
- Four runtime mechanisms operate directly on the output digit stream: exact early negative detection (END) in ReLU layers, exact sign-only decision making in the segmentation head, calibrated low-order-digit skipping, and calibrated pruning.
- A projected eight-output accelerator with shared activation delivery achieves 16.6 ms latency and 1.67 mJ per patch.
Sources (1)
- [1]DEX: Digit-Level Early Exit for Energy-Efficient MSDF Neural Network InferencearXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 11:39 AM
U-Net inference for brain-tumor segmentation requires billions of multiply-accumulate operations, motivating hardware that can reduce computation dynamically rather than relying only on fixed precision or static model compression.
Most-significant-digit-first (MSDF) arithmetic exposes the leading digits of a result during computation, enabling output-dependent decisions before the full value is generated.
Extractive summary: sentences quoted from the sources.