• Home
  • Search
  • Unified Low-rank Compression Framework for Click-through Rate Prediction
  • Open Access IconOpen Access
  • Cite Icon1
  • https://doi.org/10.1145/3637528.3671520Copy DOI Icon

Unified Low-rank Compression Framework for Click-through Rate Prediction

  • Aug 24, 2024
  • Hao Yu +4 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Deep Click-Through Rate (CTR) prediction models play an important role in modern industrial recommendation scenarios. However, high memory overhead and computational costs limit their deployment in resource-constrained environments. Low-rank approximation is an effective method for computer vision and natural language processing models, but its application in compressing CTR prediction models has been less explored. Due to the limited memory and computing resources, compression of CTR prediction models often confronts three fundamental challenges, i.e., (1). How to reduce the model sizes to adapt to edge devices? (2). How to speed up CTR prediction model inference? (3). How to retain the capabilities of original models after compression? Previous low-rank compression research mostly uses tensor decomposition, which can achieve a high parameter compression ratio, but brings in AUC degradation and additional computing overhead. To address these challenges, we propose a unified low-rank decomposition framework for compressing CTR prediction models. We find that even with the most classic matrix decomposition SVD method, our framework can achieve better performance than the original model. To further improve the effectiveness of our framework, we locally compress the output features instead of compressing the model weights. Our unified low-rank compression framework can be applied to embedding tables and MLP layers in various CTR prediction models. Extensive experiments on two academic datasets and one real industrial benchmark demonstrate that, with 3--5× model size reduction, our compressed models can achieve both faster inference and higher AUC than the uncompressed original models. Our code is at https://github.com/yuhao318/Atomic_Feature_Mimicking.

Similar Papers
  • Conference Article
  • Citations1

Embedding Normalization: Significance Preserving Feature Normalization for Click-Through Rate Prediction

  • Dec 01, 2021
  • Joonyoung Yi +2
  • Conference Article
  • Citations8

Boost CTR Prediction for New Advertisements via Modeling Visual Content

  • Dec 17, 2022
  • Tan Yu +5
  • Conference Article
  • Citations16

RLNF: Reinforcement Learning based Noise Filtering for Click-Through Rate Prediction

  • Jul 11, 2021
  • Pu Zhao +6
  • Research Article

TDAF: a bi-level optimization framework for CTR prediction with temporal drift adaptation

  • Jun 23, 2025
  • The Computer Journal
  • Guoxin Lu +2
  • Conference Article

Click-Through Rate (Ctr) Prediction for Online Advertising

  • Jan 01, 2020
  • Jessie Poh Lin Teoh +1
  • Conference Article
  • Citations3

Residual Connections Improve Prediction Performance

  • Dec 21, 2023
  • Ergun Biçici +2
  • Conference Article
  • Citations3

Feature Staleness Aware Incremental Learning for CTR Prediction

  • Aug 01, 2023
  • Zhikai Wang +3
  • Conference Article
  • Citations17

Regularized Adversarial Sampling and Deep Time-aware Attention for Click-Through Rate Prediction

  • Nov 03, 2019
  • Yikai Wang +7
  • Research Article

MLIM: A CTR prediction model describing evolution law of user interest

  • May 17, 2022
  • Web Intelligence
  • Zilong Jiang +2
  • Conference Article
  • Citations20

Looking at CTR Prediction Again: Is Attention All You Need?

  • Jul 11, 2021
  • Yuan Cheng +1
  • Research Article
  • Citations5

CowClip: Reducing CTR Prediction Model Training Time from 12 Hours to 10 Minutes on 1 GPU

  • Jun 26, 2023
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Zangwei Zheng +14
  • Research Article
  • Citations9

An ensemble learning framework for click-through rate prediction based on a reinforcement learning algorithm with parameterized actions

  • Nov 03, 2023
  • Knowledge-Based Systems
  • Mengjuan Liu +5
  • Research Article

Deep Double Towers Click Through Rate Prediction Model with Multi-Head Bilinear Fusion

  • Jan 22, 2025
  • Symmetry
  • Yuan Zhang +3
  • Conference Article
  • Citations60

MiNet

  • Oct 19, 2020
  • Wentao Ouyang +7
  • Conference Article
  • Citations9

Implicit User Awareness Modeling via Candidate Items for CTR Prediction in Search Ads

  • Apr 25, 2022
  • Kaifu Zheng +9
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.