PODNet: Pooled Outputs Distillation for Small-Tasks Incremental Learning
This addresses the problem of incremental learning for AI systems that need to accumulate knowledge over long sequences of small tasks, representing a strong specific gain in a domain-specific area.
The paper tackles catastrophic forgetting in lifelong learning by proposing PODNet, which balances remembering old classes and learning new ones, achieving accuracy gains of 12.10, 6.51, and 2.85 percentage points over state-of-the-art models on CIFAR100, ImageNet100, and ImageNet1000 datasets.
Lifelong learning has attracted much attention, but existing works still struggle to fight catastrophic forgetting and accumulate knowledge over long stretches of incremental learning. In this work, we propose PODNet, a model inspired by representation learning. By carefully balancing the compromise between remembering the old classes and learning new ones, PODNet fights catastrophic forgetting, even over very long runs of small incremental tasks --a setting so far unexplored by current works. PODNet innovates on existing art with an efficient spatial-based distillation-loss applied throughout the model and a representation comprising multiple proxy vectors for each class. We validate those innovations thoroughly, comparing PODNet with three state-of-the-art models on three datasets: CIFAR100, ImageNet100, and ImageNet1000. Our results showcase a significant advantage of PODNet over existing art, with accuracy gains of 12.10, 6.51, and 2.85 percentage points, respectively. Code is available at https://github.com/arthurdouillard/incremental_learning.pytorch