# HAWX: A Hardware-Aware FrameWork for Fast and Scalable ApproXimation of DNNs

arXiv: https://arxiv.org/abs/2602.16336 · HTML: https://mlsanity.com/paper/2602.16336/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 18 Feb 2026.

## Builds on (most cited first, 9 of 9)

- [FINN-R: An End-to-End Deep-Learning Framework for Fast Exploration of Quantized Neural Networks](https://mlsanity.com/paper/1809.04570/) (2018)
- [TFApprox: Towards a Fast Emulation of DNN Approximate Hardware Accelerators on GPU](https://mlsanity.com/paper/2002.09481/) (2020)
- [DeepAxe: A Framework for Exploration of Approximation and Reliability Trade-offs in DNN Accelerators](https://mlsanity.com/paper/2303.08226/) (2023)
- [Exploration of Activation Fault Reliability in Quantized Systolic Array-Based DNN Accelerators](https://mlsanity.com/paper/2401.09509/) (2024)
- [Special Session: Approximation and Fault Resiliency of DNN Accelerators](https://mlsanity.com/paper/2306.04645/) (2023)
- [Design Space Exploration of Approximate Computing Techniques with a Reinforcement Learning Approach](https://mlsanity.com/paper/2312.17525/) (2023)
- [ApproxDARTS: Differentiable Neural Architecture Search with Approximate Multipliers](https://mlsanity.com/paper/2404.08002/) (2024)
- [An Architectural Error Metric for CNN-Oriented Approximate Multipliers](https://mlsanity.com/paper/2408.12836/) (2024)
- [MAx-DNN: Multi-Level Arithmetic Approximation for Energy-Efficient DNN Hardware Accelerators](https://mlsanity.com/paper/2506.21371/) (2025)

## Built on by (most cited first, 2 of 2)

- [Sensitivity-Guided Framework for Pruned and Quantized Reservoir Computing Accelerators](https://mlsanity.com/paper/2603.08737/) (2026)
- [Mix-and-Match Pruning: Globally Guided Layer-Wise Sparsification of DNNs](https://mlsanity.com/paper/2603.20280/) (2026)

## Abstract (arXiv)

This work presents HAWX, a hardware-aware scalable exploration framework that employs multi-level sensitivity scoring at different DNN abstraction levels (operator, filter, layer, and model) to guide selective integration of heterogeneous AxC blocks. Supported by predictive models for accuracy, power, and area, HAWX accelerates the evaluation of candidate configurations, achieving over 23* speedup in a layer-level search with two candidate approximate blocks and more than (3*106)* speedup at the filter-level search only for LeNet-5, while maintaining accuracy comparable to exhaustive search. Experiments across state-of-the-art DNN benchmarks such as VGG-11, ResNet-18, and EfficientNetLite demonstrate that the efficiency benefits of HAWX scale exponentially with network size. The HAWX hardware-aware search algorithm supports both spatial and temporal accelerator architectures, leveraging either off-the-shelf approximate components or customized designs.