BriefGPT.xyz
May, 2019
使用卡尔曼优化器在深度神经网络中实现连续学习
Continual Learning in Deep Neural Networks by Using a Kalman Optimiser
HTML
PDF
Honglin Li, Shirin Enshaeifar, Frieder Ganz, Payam Barnaghi
TL;DR
通过使用卡尔曼优化器将神经网络分成长期和短期记忆单元，并在MNIST、CIFAR10、CIFAR100数据集上进行了评估，与基准模型进行比较显示出我们的方法使模型能够不断学习和适应新的变化，而没有忘记以前学习的任务。
Abstract
Learning and adapting to new distributions or learning new tasks sequentially without forgetting the previously learned knowledge is a challenging phenomenon in
continual learning
models. Most of the conventional
deep l
→