论文标题
终身学习的基于熵的稳定性
Entropy-based Stability-Plasticity for Lifelong Learning
论文作者
论文摘要
对于深度学习模型而言，不断学习的能力仍然难以捉摸。与人类不同，在学习新任务时，模型无法积累权重的知识，这主要是由于塑料过量过多和在训练新任务时重复使用权重的低动力。为了解决神经网络中的稳定性困境，我们提出了一种新的方法，称为基于熵的稳态性 - 塑性性（ESP）。我们的方法可以动态地决定应通过可塑性因子修改每个模型层的数量。我们将分支层和基于熵的标准纳入模型以找到这种因素。我们在自然语言和视觉领域进行的实验表明，通过减少干扰，我们的方法在利用先验知识方面的有效性。而且，在某些情况下，在训练过程中可以冻结层，从而加快训练的速度。
The ability to continuously learn remains elusive for deep learning models. Unlike humans, models cannot accumulate knowledge in their weights when learning new tasks, mainly due to an excess of plasticity and the low incentive to reuse weights when training a new task. To address the stability-plasticity dilemma in neural networks, we propose a novel method called Entropy-based Stability-Plasticity (ESP). Our approach can decide dynamically how much each model layer should be modified via a plasticity factor. We incorporate branch layers and an entropy-based criterion into the model to find such factor. Our experiments in the domains of natural language and vision show the effectiveness of our approach in leveraging prior knowledge by reducing interference. Also, in some cases, it is possible to freeze layers during training leading to speed up in training.
