论文标题
不断学习的颜色和形状表示
Disentanglement of Color and Shape Representations for Continual Learning
论文作者
论文摘要
我们假设，灾难性遗忘造成的分离特征表示形式少遭受。作为一个案例研究，我们通过调整网络体系结构来执行颜色和形状的明确分离。我们使用Oxford-102 Flowers数据集测试了分类准确性并忘记了任务收入设置。我们将方法与弹性重量巩固相结合，学习而不忘记，突触智能和记忆意识突触，并证明特征分离会对持续的学习表现产生积极影响。
We hypothesize that disentangled feature representations suffer less from catastrophic forgetting. As a case study we perform explicit disentanglement of color and shape, by adjusting the network architecture. We tested classification accuracy and forgetting in a task-incremental setting with Oxford-102 Flowers dataset. We combine our method with Elastic Weight Consolidation, Learning without Forgetting, Synaptic Intelligence and Memory Aware Synapses, and show that feature disentanglement positively impacts continual learning performance.
