Detachedly Learn a Classifier for Class-Incremental Learning
This addresses continual learning challenges for AI systems needing to adapt to new tasks without forgetting previous ones, representing an incremental improvement.
The paper tackles the problem of knowledge degradation and prediction bias in class-incremental learning by proposing a task-aware experience replay strategy that detaches old task classifiers during updates, resulting in outperforming current state-of-the-art methods.
In continual learning, model needs to continually learn a feature extractor and classifier on a sequence of tasks. This paper focuses on how to learn a classifier based on a pretrained feature extractor under continual learning setting. We present an probabilistic analysis that the failure of vanilla experience replay (ER) comes from unnecessary re-learning of previous tasks and incompetence to distinguish current task from the previous ones, which is the cause of knowledge degradation and prediction bias. To overcome these weaknesses, we propose a novel replay strategy task-aware experience replay. It rebalances the replay loss and detaches classifier weight for the old tasks from the update process, by which the previous knowledge is kept intact and the overfitting on episodic memory is alleviated. Experimental results show our method outperforms current state-of-the-art methods.