# Explainable AI-Guided Efficient Approximate DNN Generation for Multi-Pod Systolic Arrays

arXiv: https://arxiv.org/abs/2503.16583 · HTML: https://mlsanity.com/paper/2503.16583/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 28 references that are in the map.
- First posted on arXiv 20 Mar 2025.

## Builds on (most cited first, 19 of 19)

- [Captum: A unified and generic model interpretability library for PyTorch](https://mlsanity.com/paper/2009.07896/) (2020)
- [The many Shapley values for model explanation](https://mlsanity.com/paper/1908.08474/) (2019)
- [TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings](https://mlsanity.com/paper/2304.01433/) (2023)
- [Pruning by Explaining: A Novel Criterion for Deep Neural Network Pruning](https://mlsanity.com/paper/1912.08881/) (2019)
- [How Important Is a Neuron?](https://mlsanity.com/paper/1805.12233/) (2018)
- [Hardware Approximate Techniques for Deep Neural Network Accelerators: A Survey](https://mlsanity.com/paper/2203.08737/) (2022)
- [Evaluating Explanation Methods for Deep Learning in Security](https://mlsanity.com/paper/1906.02108/) (2019)
- [Hardware and Software Optimizations for Accelerating Deep Neural Networks: Survey of Current Trends, Challenges, and the Road Ahead](https://mlsanity.com/paper/2012.11233/) (2020)
- [ALWANN: Automatic Layer-Wise Approximation of Deep Neural Network Accelerators without Retraining](https://mlsanity.com/paper/1907.07229/) (2019)
- [Utilizing Explainable AI for Quantization and Pruning of Deep Neural Networks](https://mlsanity.com/paper/2008.09072/) (2020)
- [Capsule Network is Not More Robust than Convolutional Network](https://mlsanity.com/paper/2103.15459/) (2021)
- [Control Variate Approximation for DNN Accelerators](https://mlsanity.com/paper/2102.09642/) (2021)
- [AdaPT: Fast Emulation of Approximate DNN Accelerators in PyTorch](https://mlsanity.com/paper/2203.04071/) (2022)
- [ReD-CaNe: A Systematic Methodology for Resilience Analysis and Design of Capsule Networks under Approximations](https://mlsanity.com/paper/1912.00700/) (2019)
- [Exploring Fault-Energy Trade-offs in Approximate DNN Hardware Accelerators](https://mlsanity.com/paper/2101.02860/) (2021)
- [Scale-out Systolic Arrays](https://mlsanity.com/paper/2203.11540/) (2022)
- [Exposing Reliability Degradation and Mitigation in Approximate DNNs under Permanent Faults](https://mlsanity.com/paper/2301.07484/) (2023)
- [Is Approximation Universally Defensive Against Adversarial Attacks in Deep Neural Networks?](https://mlsanity.com/paper/2112.01555/) (2021)
- [Evolutionary Neural Architecture Search Supporting Approximate Multipliers](https://mlsanity.com/paper/2101.11883/) (2021)

## Abstract (arXiv)

Approximate deep neural networks (AxDNNs) are promising for enhancing energy efficiency in real-world devices. One of the key contributors behind this enhanced energy efficiency in AxDNNs is the use of approximate multipliers. Unfortunately, the simulation of approximate multipliers does not usually scale well on CPUs and GPUs. As a consequence, this slows down the overall simulation of AxDNNs aimed at identifying the appropriate approximate multipliers to achieve high energy efficiency with a minimum accuracy loss. To address this problem, we present a novel XAI-Gen methodology, which leverages the analytical model of the emerging hardware accelerator (e.g., Google TPU v4) and explainable artificial intelligence (XAI) to precisely identify the non-critical layers for approximation and quickly discover the appropriate approximate multipliers for AxDNN layers. Our results show that XAI-Gen achieves up to 7x lower energy consumption with only 1-2% accuracy loss. We also showcase the effectiveness of the XAI-Gen approach through a neural architecture search (XAI-NAS) case study. Interestingly, XAI-NAS achieves 40% higher energy efficiency with up to 5x less execution time when compared to the state-of-the-art NAS methods for generating AxDNNs.