Pseudo-Recursal: Solving the Catastrophic Forgetting Problem in Deep Neural Networks
This addresses the problem of sequential learning for AI researchers, offering a substantial improvement over current state-of-the-art solutions.
The paper tackles catastrophic forgetting in deep neural networks by using a Generative Adversarial Network to generate items for pseudo-rehearsal, enabling sequential learning on CIFAR-10, SVHN, and MNIST with only a 1.67% absolute accuracy loss on CIFAR-10 and a 0.24% gain on SVHN.
In general, neural networks are not currently capable of learning tasks in a sequential fashion. When a novel, unrelated task is learnt by a neural network, it substantially forgets how to solve previously learnt tasks. One of the original solutions to this problem is pseudo-rehearsal, which involves learning the new task while rehearsing generated items representative of the previous task/s. This is very effective for simple tasks. However, pseudo-rehearsal has not yet been successfully applied to very complex tasks because in these tasks it is difficult to generate representative items. We accomplish pseudo-rehearsal by using a Generative Adversarial Network to generate items so that our deep network can learn to sequentially classify the CIFAR-10, SVHN and MNIST datasets. After training on all tasks, our network loses only 1.67% absolute accuracy on CIFAR-10 and gains 0.24% absolute accuracy on SVHN. Our model's performance is a substantial improvement compared to the current state of the art solution.