Efficient Source-Free Time-Series Adaptation via Parameter Subspace Disentanglement
Patel, Gaurav, Sandino, Christopher, Mahasseni, Behrooz, Zippi, Ellen L, Azemi, Erdrin, Moin, Ali, Minxha, Juri
–arXiv.org Artificial Intelligence
In this paper, we propose a framework for efficient Source-Free Domain Adaptation (SFDA) in the context of time-series, focusing on enhancing both parameter efficiency and data-sample utilization. Our approach introduces an improved paradigm for source-model preparation and target-side adaptation, aiming to enhance training efficiency during target adaptation. Specifically, we reparameterize the source model's weights in a Tucker-style decomposed manner, factorizing the model into a compact form during the source model preparation phase. During target-side adaptation, only a subset of these decomposed factors is fine-tuned, leading to significant improvements in training efficiency. We demonstrate using PAC Bayesian analysis that this selective fine-tuning strategy implicitly regularizes the adaptation process by constraining the model's learning capacity. Furthermore, this re-parameterization reduces the overall model size and enhances inference efficiency, making the approach particularly well suited for resource-constrained devices. Additionally, we demonstrate that our framework is compatible with various SFDA methods and achieves significant computational efficiency, reducing the number of fine-tuned parameters and inference overhead in terms of MACs by over 90% while maintaining model performance.
arXiv.org Artificial Intelligence
Oct-2-2024
- Country:
- Asia (0.14)
- Genre:
- Research Report > New Finding (0.92)
- Industry:
- Health & Medicine (0.67)
- Technology: