# ADC/DAC-Free Analog Acceleration of Deep Neural Networks with Frequency Transformation

arXiv: https://arxiv.org/abs/2309.01771 · HTML: https://mlsanity.com/paper/2309.01771/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 58 references that are in the map.
- First posted on arXiv 4 Sep 2023.

## Builds on (most cited first, 21 of 21)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding](https://mlsanity.com/paper/1510.00149/) (2015)
- [Rethinking the Value of Network Pruning](https://mlsanity.com/paper/1810.05270/) (2018)
- [Learning Structured Sparsity in Deep Neural Networks](https://mlsanity.com/paper/1608.03665/) (2016)
- [Pruning neural networks without any data by iteratively conserving synaptic flow](https://mlsanity.com/paper/2006.05467/) (2020)
- [Frequency Principle: Fourier Analysis Sheds Light on Deep Neural Networks](https://mlsanity.com/paper/1901.06523/) (2019)
- [Learning in the Frequency Domain](https://mlsanity.com/paper/2002.12416/) (2020)
- [Structured Pruning of Deep Convolutional Neural Networks](https://mlsanity.com/paper/1512.08571/) (2015)
- [Compressing Convolutional Neural Networks](https://mlsanity.com/paper/1506.04449/) (2015)
- [Fast Walsh-Hadamard Transform and Smooth-Thresholding Based Binary Layers in Deep Neural Networks](https://mlsanity.com/paper/2104.07085/) (2021)
- [Block Walsh-Hadamard Transform Based Binary Layers in Deep Neural Networks](https://mlsanity.com/paper/2201.02711/) (2022)
- [MF-Net: Compute-In-Memory SRAM for Multibit Precision Inference using Memory-immersed Data Conversion and Multiplication-free Operators](https://mlsanity.com/paper/2102.00035/) (2021)
- [A Hybrid Quantum-Classical Approach based on the Hadamard Transform for the Convolutional Layer](https://mlsanity.com/paper/2305.17510/) (2023)
- [Walsh-Hadamard Variational Inference for Bayesian Deep Learning](https://mlsanity.com/paper/1905.11248/) (2019)
- [MC-CIM: Compute-in-Memory with Monte-Carlo Dropouts for Bayesian Edge Intelligence](https://mlsanity.com/paper/2111.07125/) (2021)
- [DCT Perceptron Layer: A Transform Domain Approach for Convolution Layer](https://mlsanity.com/paper/2211.08577/) (2022)
- [Robust Monocular Localization of Drones by Adapting Domain Maps to Depth Prediction Inaccuracies](https://mlsanity.com/paper/2210.15559/) (2022)
- [A hybrid machine learning framework for clad characteristics prediction in metal additive manufacturing](https://mlsanity.com/paper/2307.01872/) (2023)
- [Memory-Immersed Collaborative Digitization for Area-Efficient Compute-in-Memory Deep Learning](https://mlsanity.com/paper/2307.03863/) (2023)

## Built on by (most cited first, 3 of 3)

- [Intelligent Sensing-to-Action for Robust Autonomy at the Edge: Opportunities and Challenges](https://mlsanity.com/paper/2502.02692/) (2025)
- [Neural Precision Polarization: Simplifying Neural Network Inference with Dual-Level Precision](https://mlsanity.com/paper/2411.05845/) (2024)
- [Containing Analog Data Deluge at Edge through Frequency-Domain Compression in Collaborative Compute-in-Memory Networks](https://mlsanity.com/paper/2309.11048/) (2023)

## Abstract (arXiv)

The edge processing of deep neural networks (DNNs) is becoming increasingly important due to its ability to extract valuable information directly at the data source to minimize latency and energy consumption. Frequency-domain model compression, such as with the Walsh-Hadamard transform (WHT), has been identified as an efficient alternative. However, the benefits of frequency-domain processing are often offset by the increased multiply-accumulate (MAC) operations required. This paper proposes a novel approach to an energy-efficient acceleration of frequency-domain neural networks by utilizing analog-domain frequency-based tensor transformations. Our approach offers unique opportunities to enhance computational efficiency, resulting in several high-level advantages, including array micro-architecture with parallelism, ADC/DAC-free analog computations, and increased output sparsity. Our approach achieves more compact cells by eliminating the need for trainable parameters in the transformation matrix. Moreover, our novel array micro-architecture enables adaptive stitching of cells column-wise and row-wise, thereby facilitating perfect parallelism in computations. Additionally, our scheme enables ADC/DAC-free computations by training against highly quantized matrix-vector products, leveraging the parameter-free nature of matrix multiplications. Another crucial aspect of our design is its ability to handle signed-bit processing for frequency-based transformations. This leads to increased output sparsity and reduced digitization workload. On a 16×16 crossbars, for 8-bit input processing, the proposed approach achieves the energy efficiency of 1602 tera operations per second per Watt (TOPS/W) without early termination strategy and 5311 TOPS/W with early termination strategy at VDD = 0.8 V.