Making Models Shallow Again: Jointly Learning to Reduce Non-Linearity and Depth for Latency-Efficient Private Inference

Souvik Kundu, Yuke Zhang, Dake Chen, Peter A. Beerel
4/26/2023
cs.LGcs.CR

No confident code implementation found for this paper.

Abstract

Large number of ReLU and MAC operations of Deep neural networks make them ill-suited for latency and compute-efficient private inference. In this paper, we present a model optimization method that allows a model to learn to be shallow. In particular, we leverage the ReLU sensitivity of a convolutional block to remove a ReLU layer and merge its succeeding and preceding convolution layers to a shallow block. Unlike existing ReLU reduction methods, our joint reduction method can yield models with improved reduction of both ReLUs and linear operations by up to 1.73x and 1.47x, respectively, evaluated with ResNet18 on CIFAR-100 without any significant accuracy-drop.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Learning to Linearize Deep Neural Networks for Secure and Efficient Private Inference

88% match2023

Optimally Deep Networks -- Adapting Model Depth to Datasets for Superior Efficiency

83% match2025

Efficient Latency-Aware CNN Depth Compression via Two-Stage Dynamic Programming

82% match1 repo132023

Make Deep Networks Shallow Again

81% match2023

Principled Approximation Methods for Efficient and Scalable Deep Learning

81% match2025

ANCRe: Adaptive Neural Connection Reassignment for Efficient Depth Scaling

81% match2026

Cite this paper

@article{kundu2023making,
  title  = {Making Models Shallow Again: Jointly Learning to Reduce Non-Linearity and Depth for Latency-Efficient Private Inference},
  author = {Souvik Kundu and Yuke Zhang and Dake Chen and Peter A. Beerel},
  year   = {2023},
  eprint = {2304.13274},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2304.13274v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2304.13274.svg)](https://codeofpaper.com/paper/2304.13274)

Discussion