Randomized Asymmetric Chain of LoRA: The First Meaningful Theoretical Framework for Low-Rank Adaptation
About
Fine-tuning has become a popular approach to adapting large foundational models to specific tasks. As the size of models and datasets grows, parameter-efficient fine-tuning techniques are increasingly important. One of the most widely used methods is Low-Rank Adaptation (LoRA), with adaptation update expressed as the product of two low-rank matrices. While LoRA was shown to possess strong performance in fine-tuning, it often under-performs when compared to full-parameter fine-tuning (FPFT). Although many variants of LoRA have been extensively studied empirically, their theoretical optimization analysis is heavily under-explored. The starting point of our work is a demonstration that LoRA and its two extensions, Asymmetric LoRA and Chain of LoRA, indeed encounter convergence issues. To address these issues, we propose Randomized Asymmetric Chain of LoRA (RAC-LoRA) -- a general optimization framework that rigorously analyzes the convergence rates of LoRA-based methods. Our approach inherits the empirical benefits of LoRA-style heuristics, but introduces several small but important algorithmic modifications which turn it into a provably convergent method. Our framework serves as a bridge between FPFT and low-rank adaptation. We provide provable guarantees of convergence to the same solution as FPFT, along with the rate of convergence. Additionally, we present a convergence analysis for smooth, non-convex loss functions, covering gradient descent, stochastic gradient descent, and federated learning settings. Our theoretical findings are supported by experimental results.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Image Classification | Office-Home (test) | Mean Accuracy89.33 | 402 | |
| Image Classification | OfficeHome | -- | 161 | |
| Paraphrase Detection | MRPC | Empirical Error G(W)0.37 | 90 | |
| Relational Reasoning | CLUTRR | Empirical Generalization Error G(W)1.01 | 60 | |
| Linguistic Acceptability | CoLA (test) | Avg Accuracy68.1 | 56 | |
| Linguistic Acceptability | COLA | Generalization Error Approximation (G(W))0.25 | 40 | |
| Image Classification | OfficeHome | Generalization Error (G(W))0.1 | 40 | |
| Paraphrase Identification | PAWS | G(W) Error Approximation1.27 | 40 | |
| Question Classification | TREC50 | Generalization Error G(W)2.24 | 30 | |
| Image Classification | CIFAR-10 (test) | Test Accuracy99 | 24 |