论文标题
伪造的伪级用于持续学习和正常流量
Pseudo-Rehearsal for Continual Learning with Normalizing Flows
论文作者
论文摘要
每当神经网络覆盖过去的知识同时接受新任务培训时，就会发生灾难性的遗忘（CF）。处理CF的常见技术包括权重的正规化（例如，使用它们在过去的任务上的重要性）和排练策略，在这些策略中，网络在过去的数据上不断重新训练。为了具有无尽的数据源，也已将生成模型应用于后者。在本文中，我们提出了一种新颖的方法，该方法结合了正则化和基于生成的彩排方法的优势。我们的生成模型由概率和可逆神经网络的标准化流（NF）组成，该网络对网络的内部嵌入进行了训练。通过在任务上保持单个NF条件，我们表明我们的内存开销保持恒定。此外，利用NF的可逆性，我们提出了一种简单的方法，以使网络对过去任务的嵌入正规化。我们表明，我们的方法在文献中具有有限的计算能力和内存开销方面的最新方法方面表现出色。
Catastrophic forgetting (CF) happens whenever a neural network overwrites past knowledge while being trained on new tasks. Common techniques to handle CF include regularization of the weights (using, e.g., their importance on past tasks), and rehearsal strategies, where the network is constantly re-trained on past data. Generative models have also been applied for the latter, in order to have endless sources of data. In this paper, we propose a novel method that combines the strengths of regularization and generative-based rehearsal approaches. Our generative model consists of a normalizing flow (NF), a probabilistic and invertible neural network, trained on the internal embeddings of the network. By keeping a single NF conditioned on the task, we show that our memory overhead remains constant. In addition, exploiting the invertibility of the NF, we propose a simple approach to regularize the network's embeddings with respect to past tasks. We show that our method performs favorably with respect to state-of-the-art approaches in the literature, with bounded computational power and memory overheads.
