skip to main content
Idioma:

AUTOTRAINER: An Automatic DNN Training Problem Detection and Repair System

Zhang, Xiaoyu ; Zhai, Juan ; Ma, Shiqing ; Shen, Chao

2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE), 2021, p.359-371

IEEE

Texto completo disponível

Citações Citado por
  • Título:
    AUTOTRAINER: An Automatic DNN Training Problem Detection and Repair System
  • Autor: Zhang, Xiaoyu ; Zhai, Juan ; Ma, Shiqing ; Shen, Chao
  • Assuntos: deep learning training ; Maintenance engineering ; Monitoring ; Recurrent neural networks ; Software ; Software engineering ; software tools ; Training
  • É parte de: 2021 IEEE/ACM 43rd International Conference on Software Engineering (ICSE), 2021, p.359-371
  • Descrição: With machine learning models especially Deep Neural Network (DNN) models becoming an integral part of the new intelligent software, new tools to support their engineering process are in high demand. Existing DNN debugging tools are either post-training which wastes a lot of time training a buggy model and requires expertises, or limited on collecting training logs without analyzing the problem not even fixing them. In this paper, we propose AUTOTRAINER, a DNN training monitoring and automatic repairing tool which supports detecting and auto repairing five commonly seen training problems. During training, it periodically checks the training status and detects potential problems. Once a problem is found, AUTOTRAINER tries to fix it by using built-in state-of-the-art solutions. It supports various model structures and input data types, such as Convolutional Neural Networks (CNNs) for image and Recurrent Neural Networks (RNNs) for texts. Our evaluation on 6 datasets, 495 models show that AUTOTRAINER can effectively detect all potential problems with 100% detection rate and no false positives. Among all models with problems, it can fix 97.33% of them, increasing the accuracy by 47.08% on average.
  • Editor: IEEE
  • Idioma: Inglês
Voltar para lista de resultados

  • Realização: ABCD-USP USP   Logos de Redes Sociais: Freepik

Buscando em bases de dados remotas. Favor aguardar.