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 1902.09432cs.LGstat.ML
收藏

基于可加性参数分解的可扩展有序鲁棒持续学习

Scalable and Order-robust Continual Learning with Additive Parameter Decomposition

KAIST(韩国科学技术院);AITRICS

Jaehong Yoon, Saehoon Kim, Eunho Yang, Sung Ju Hwang

摘要:

虽然最近的持续学习方法在很大程度上缓解了玩具大小的数据集上的灾难性问题,但仍有一些问题有待解决,以将其应用于现实世界的问题领域。首先,持续学习模型应该有效地处理灾难性遗忘,并且即使有大量任务也能有效地进行训练。其次,它需要解决顺序敏感性问题,其中任务的性能根据任务到达序列的顺序而变化很大,因为它可能会导致公平性起关键作用的严重问题(例如医疗诊断)。为了解决这些实际挑战,我们提出了一种新的持续学习方法,该方法是可扩展的,以及顺序鲁棒性,而不是学习一个完全共享的权重集,表示每个任务的参数作为任务共享和稀疏任务自适应参数的总和。使用我们的加性参数分解(APD),早期任务的任务自适应参数基本上不受影响,我们只更新它们以反映对任务共享参数所做的更改。参数的这种分解有效地防止了灾难性遗忘和阶次敏感性,同时具有计算和存储效率。此外,我们可以实现更好的可扩展性与APD使用分层知识整合,集群的任务自适应参数,以获得分层共享的参数。我们在多个基准数据集上使用APD,APD-Net对最先进的持续学习方法验证了我们的网络,该方法在准确性,可扩展性和顺序鲁棒性方面表现出色。

英文摘要:

While recent continual learning methods largely alleviate the catastrophic problem on toy-sized datasets, some issues remain to be tackled to apply them to real-world problem domains. First, a continual learning model should effectively handle catastrophic forgetting and be efficient to train even with a large number of tasks. Secondly, it needs to tackle the problem of order-sensitivity, where the performance of the tasks largely varies based on the order of the task arrival sequence, as it may cause serious problems where fairness plays a critical role (e.g. medical diagnosis). To tackle these practical challenges, we propose a novel continual learning method that is scalable as well as order-robust, which instead of learning a completely shared set of weights, represents the parameters for each task as a sum of task-shared and sparse task-adaptive parameters. With our Additive Parameter Decomposition (APD), the task-adaptive parameters for earlier tasks remain mostly unaffected, where we update them only to reflect the changes made to the task-shared parameters. This decomposition of parameters effectively prevents catastrophic forgetting and order-sensitivity, while being computation- and memory-efficient. Further, we can achieve even better scalability with APD using hierarchical knowledge consolidation, which clusters the task-adaptive parameters to obtain hierarchically shared parameters. We validate our network with APD, APD-Net, on multiple benchmark datasets against state-of-the-art continual learning methods, which it largely outperforms in accuracy, scalability, and order-robustness.

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。

↑