arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1908.11033cs.LGcs.AIstat.ML
收藏

基于GBDT的终身学习自动机器学习框架

An Auto-ML Framework Based on GBDT for Lifelong Learning

Beijing University of Posts and Telecommunications(北京邮电大学);Central South University(中南大学);Inspur Electronic Information Industry Co.,Ltd(浪潮电子信息产业股份有限公司)

Jinlong Chai, Jiangeng Chang, Yakun Zhao, Honggang Liu

摘要:

自动机器学习(Auto-ML)近年来受到越来越多的关注,我们的工作就是解决数据漂移问题,即数据的分布会随着采集过程逐渐发生变化,导致Auto-ML模型的性能变差。我们基于GBDT构建模型,使用增量学习和完全学习来处理漂移问题。实验表明,我们的方法表现良好的五个数据集。结果表明,该方法能有效地解决数据漂移问题,具有较强的鲁棒性。

英文摘要:

Automatic Machine Learning (Auto-ML) has attracted more and more attention in recent years, our work is to solve the problem of data drift, which means that the distribution of data will gradually change with the acquisition process, resulting in a worse performance of the auto-ML model. We construct our model based on GBDT, Incremental learning and full learning are used to handle with drift problem. Experiments show that our method performs well on the five data sets. Which shows that our method can effectively solve the problem of data drift and has robust performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑