arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1906.02425cs.LGcs.AIcs.CVstat.ML
收藏

基于贝叶斯神经网络的不确定性引导持续学习

Uncertainty-guided Continual Learning with Bayesian Neural Networks

UC Berkeley(加州大学伯克利分校);Facebook AI Research(脸书人工智能研究院);KAUST(阿卜杜拉国王科技大学);Stanford University(斯坦福大学)

Sayna Ebrahimi, Mohamed Elhoseiny, Trevor Darrell, Marcus Rohrbach

摘要:

持续学习的目的是学习新的任务而不忘记以前学过的。当无法访问先前任务的数据以及模型具有固定容量时,这尤其具有挑战性。目前基于正则化的持续学习算法需要一个外部表示和额外的计算来测量参数的\textit{importance}。相比之下,我们提出了不确定性指导的连续贝叶斯神经网络(UCB),其中学习率根据网络中权重的概率分布中定义的不确定性进行调整。不确定性是一种自然的方式来识别\textit{what to remember}和\textit{what to change},因为我们不断学习,从而减轻灾难性遗忘。我们还展示了我们模型的一个变体,它使用不确定性进行权重修剪,并通过保存每个任务的二进制掩码来修剪后保留任务性能。我们在不同的对象分类数据集上广泛评估了我们的UCB方法,这些数据集具有短序列和长序列的任务,并报告了与现有方法相比的优越或同等的性能。此外,我们表明,我们的模型并不一定需要在测试时的任务信息,即它不假定知识的任务样本属于。

英文摘要:

Continual learning aims to learn new tasks without forgetting previously learned ones. This is especially challenging when one cannot access data from previous tasks and when the model has a fixed capacity. Current regularization-based continual learning algorithms need an external representation and extra computation to measure the parameters' \textit{importance}. In contrast, we propose Uncertainty-guided Continual Bayesian Neural Networks (UCB), where the learning rate adapts according to the uncertainty defined in the probability distribution of the weights in networks. Uncertainty is a natural way to identify \textit{what to remember} and \textit{what to change} as we continually learn, and thus mitigate catastrophic forgetting. We also show a variant of our model, which uses uncertainty for weight pruning and retains task performance after pruning by saving binary masks per tasks. We evaluate our UCB approach extensively on diverse object classification datasets with short and long sequences of tasks and report superior or on-par performance compared to existing approaches. Additionally, we show that our model does not necessarily need task information at test time, i.e. it does not presume knowledge of which task a sample belongs to.

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。

↑