paper-with-me

홈 › Papers

Accelerating Recommender Model Training by Dynamically Skipping Stale Embeddings

2024-03-22 · Yassaman Ebrahimzadeh Maboud, Muhammad Adnan, Divya Mahajan, Prashant J. Nair

Training recommendation models pose significant challenges regarding resource utilization and performance. Prior research has proposed an approach that categorizes embeddings into popular and non-popular classes to reduce the training time for recommendation models. We observe that, even among the popular embeddings, certain embeddings undergo rapid training and exhibit minimal subsequent variation, resulting in saturation. Consequently, updates to these embeddings lack any contribution to model quality. This paper presents Slipstream, a software framework that identifies stale embeddings on the fly and skips their updates to enhance performance. This capability enables Slipstream to achieve substantial speedup, optimize CPU-GPU bandwidth usage, and eliminate unnecessary memory access. SlipStream showcases training time reductions of 2x, 2.4x, 1.2x, and 1.175x across real-world datasets and configurations, compared to Baseline XDL, Intel-optimized DRLM, FAE, and Hotline, respectively.

📄 PDF Abstract BibTeX arXiv:2404.04270

Code (0)

등록된 구현이 없습니다.

Tasks

CPUGPU

Similar Papers 제목 키워드 기반

VVS: Accelerating Speculative Decoding for Visual Autoregressive Generation via Partial Verification Skipping

2025-11-17 · Haotian Dong, Ye Li, Rongwei Lu, Chen Tang 외 arxiv

Visual autoregressive (AR) generation models have demonstrated strong potential for image generation, yet their next-token-prediction paradigm introduces considerable inference latency. Although speculative decoding (SD)…

Image Generation

Accelerating Deep Learning by Focusing on the Biggest Losers

2019-10-02 · Angela H. Jiang, Daniel L. -K. Wong, Giulio Zhou, David G. Andersen 외

This paper introduces Selective-Backprop, a technique that accelerates the training of deep neural networks (DNNs) by prioritizing examples with high loss at each iteration. Selective-Backprop uses the output of a traini…

Deep Learning

CLaSp: In-Context Layer Skip for Self-Speculative Decoding

2025-05-30 · Longze Chen, Renke Shan, Huiming Wang, Lu Wang 외

Speculative decoding (SD) is a promising method for accelerating the decoding process of Large Language Models (LLMs). The efficiency of SD primarily hinges on the consistency between the draft model and the verify model…

Feature Staleness Aware Incremental Learning for CTR Prediction

2025-04-29 · Zhikai Wang, Yanyan Shen, Zibin Zhang, Kangyi Lin

Click-through Rate (CTR) prediction in real-world recommender systems often deals with billions of user interactions every day. To improve the training efficiency, it is common to update the CTR prediction model incremen…

Click-Through Rate PredictionIncremental LearningPredictionRecommendation Systems

Accelerating Distributed ML Training via Selective Synchronization

2023-07-16 · Sahil Tyagi, Martin Swany

In distributed training, deep neural networks (DNNs) are launched over multiple workers concurrently and aggregate their local updates on each step in bulk-synchronous parallel (BSP) training. However, BSP does not linea…