LACE: Loss-Adaptive Capacity Expansion for Continual Learning
About
Fixed representational capacity is a fundamental constraint in continual learning: practitioners must guess an appropriate model width before training, without knowing how many distinct concepts the data contains. We propose LACE (Loss-Adaptive Capacity Expansion), a simple online mechanism that expands a model's representational capacity during training by monitoring its own loss signal. When sustained loss deviation exceeds a threshold - indicating that the current capacity is insufficient for newly encountered data - LACE adds new dimensions to the projection layer and trains them jointly with existing parameters. Across synthetic and real-data experiments, LACE triggers expansions exclusively at domain boundaries (100% boundary precision, zero false positives), matches the accuracy of a large fixed-capacity model while starting from a fraction of its dimensions, and produces adapter dimensions that are collectively critical to performance (3% accuracy drop when all adapters removed). We further demonstrate unsupervised domain separation in GPT-2 activations via layer-wise clustering, showing a U-shaped separability curve across layers that motivates adaptive capacity allocation in deep networks. LACE requires no labels, no replay buffers, and no external controllers, making it suitable for on-device continual learning under resource constraints.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Classification | 10 Synthetic Domains | Accuracy99.9 | 3 | |
| Domain Classification | 50 domains 10 families x 5 variants | Accuracy67.6 | 3 | |
| Sequential Domain Classification | Wikipedia -> Code -> Chat | Accuracy79.6 | 3 |