Continual Learning with Synthetic Boundary Experience Blending
This addresses the problem of fragile decision boundaries in continual learning for AI systems, offering a novel method to enhance stability and robustness, though it is incremental as it builds on experience replay.
The paper tackled catastrophic forgetting in continual learning by introducing synthetic boundary data generated via differential privacy-inspired noise, which enriches feature space near decision boundaries, resulting in accuracy improvements of 10%, 6%, and 13% on CIFAR-10, CIFAR-100, and Tiny ImageNet over baselines.
Continual learning (CL) seeks to mitigate catastrophic forgetting when models are trained with sequential tasks. A common approach, experience replay (ER), stores past exemplars but only sparsely approximates the data distribution, yielding fragile and oversimplified decision boundaries. We address this limitation by introducing synthetic boundary data (SBD), generated via differential privacy: inspired noise into latent features to create boundary-adjacent representations that implicitly regularize decision boundaries. Building on this idea, we propose Experience Blending (EB), a framework that jointly trains on exemplars and SBD through a dual-model aggregation strategy. EB has two components: (1) latent-space noise injection to synthesize boundary data, and (2) end-to-end training that jointly leverages exemplars and SBD. Unlike standard experience replay, SBD enriches the feature space near decision boundaries, leading to more stable and robust continual learning. Extensive experiments on CIFAR-10, CIFAR-100, and Tiny ImageNet demonstrate consistent accuracy improvements of 10%, 6%, and 13%, respectively, over strong baselines.