Learning to Continually Learn Rapidly from Few and Noisy Data
This incremental improvement addresses continual learning challenges for AI systems dealing with limited memory and noisy data.
The paper tackled catastrophic forgetting in neural networks by combining replay mechanics with meta-learning to learn per-parameter learning rates for past tasks, resulting in strong performance with less memory, robustness to noise, and higher accuracy in fewer updates.
Neural networks suffer from catastrophic forgetting and are unable to sequentially learn new tasks without guaranteed stationarity in data distribution. Continual learning could be achieved via replay -- by concurrently training externally stored old data while learning a new task. However, replay becomes less effective when each past task is allocated with less memory. To overcome this difficulty, we supplemented replay mechanics with meta-learning for rapid knowledge acquisition. By employing a meta-learner, which \textit{learns a learning rate per parameter per past task}, we found that base learners produced strong results when less memory was available. Additionally, our approach inherited several meta-learning advantages for continual learning: it demonstrated strong robustness to continually learn under the presence of noises and yielded base learners to higher accuracy in less updates.