基于LoRA的持续学习：对关键参数变化施加约束
LoRA-Based Continual Learning with Constraints on Critical Parameter Changes
AI总结：
针对正交LoRA调优下预任务关键参数在学习后任务时仍显著变化的问题，提出冻结ViT中预任务关键参数矩阵，并基于QR分解提出正交LoRA组合（LoRAC）方法，在多个基准上实现SOTA性能。
摘要：
基于LoRA的持续学习是在下游持续学习任务中利用预训练模型的一个有前景的方向。近期研究表明，正交LoRA调优能有效缓解遗忘问题。然而，本研究揭示，在正交LoRA调优下，预任务的关键参数在学习后任务后仍会发生显著变化。为解决这一问题，我们提出在学习后任务前直接冻结Vision Transformer（ViT）中对预任务最关键的参数矩阵。此外，基于正交LoRA调优，我们提出了一种基于QR分解的正交LoRA组合（LoRAC）方法，这可能进一步增强我们方法的可塑性。详尽的消融研究和广泛的对比实验证明了所提方法的有效性。结果表明，我们的方法在多个知名持续学习基准上达到了最先进（SOTA）性能。例如，在Split CIFAR-100数据集上，与之前的方法相比，我们的方法准确率提升了6.35%，遗忘率降低了3.24%。我们的代码可在https://github.com/learninginvision/LoRAC-IPC获取。
英文摘要：
LoRA-based continual learning represents a promising avenue for leveraging pre-trained models in downstream continual learning tasks. Recent studies have shown that orthogonal LoRA tuning effectively mitigates forgetting. However, this work unveils that under orthogonal LoRA tuning, the critical parameters for pre-tasks still change notably after learning post-tasks. To address this problem, we directly propose freezing the most critical parameter matrices in the Vision Transformer (ViT) for pre-tasks before learning post-tasks. In addition, building on orthogonal LoRA tuning, we propose orthogonal LoRA composition (LoRAC) based on QR decomposition, which may further enhance the plasticity of our method. Elaborate ablation studies and extensive comparisons demonstrate the effectiveness of our proposed method. Our results indicate that our method achieves state-of-the-art (SOTA) performance on several well-known continual learning benchmarks. For instance, on the Split CIFAR-100 dataset, our method shows a 6.35\% improvement in accuracy and a 3.24\% reduction in forgetting compared to previous methods. Our code is available at https://github.com/learninginvision/LoRAC-IPC.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
