LoRA-RC: Reservoir Computing with Low-Rank Adaptation
Abstract
Reservoir computing (RC) trains only a linear readout over a fixed recurrent layer, making it fast and data-efficient for online prediction. However, a static reservoir degrades under system drift, readout-only adaptation is then insufficient, and unconstrained reservoir adaptation can destroy the echo-state and incremental stability properties that make RC reliable. This paper proposes LoRA-RC, which adapts the recurrent matrix through a low-rank correction driven by streaming prediction errors. The base reservoir and adaptation bases are fixed offline; a small core matrix is adapted online, projected onto a spectral-norm ball, and low-pass filtered at each step. The projection guarantees that every applied recurrent matrix remains within a certified contraction set, and an incremental input-to-state stability bound is established for the reservoir along each online adaptation path, with path-independent rate and gain. On a Lorenz system with an abrupt parameter drift, LoRA-RC cuts post-drift prediction error by 56% versus a fixed RC and 51% versus readout-only adaptation; ablations over 20 seeds show that removing the projection inflates this error by more than a factor of 40.
Related
- From What Is Used to What Is Missing: Rethinking Dynamic Rank in LoRA
- Rank-Efficient LoRA via Joint Tangent-Space Optimization under Isotropic Curvature
- The internal model principle in reservoir computing
- A hidden bottleneck in classical and quantum linear reservoir computing
- Uncertainty Quantification of Next Generation Reservoir Computing with Applications to Memory-Driven Dynamical Systems
BibTeX
@misc{wan2026lorarc,
title = {{LoRA-RC: Reservoir Computing with Low-Rank Adaptation}},
author = {Wenbin Wan},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.12327},
eprint = {2609.12327},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.12327},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - LoRA-RC: Reservoir Computing with Low-Rank Adaptation
AU - Wenbin Wan
PY - 2026
DA - 2026/09/11
PB - arXiv
DO - 10.48550/arxiv.2609.12327
UR - https://arxiv.org/abs/2609.12327
AB - Reservoir computing (RC) trains only a linear readout over a fixed recurrent layer, making it fast and data-efficient for online prediction. However, a static reservoir degrades under system drift, readout-only adaptation is then insufficient, and unconstrained reservoir adaptation can destroy the echo-state and incremental stability properties that make RC reliable. This paper proposes LoRA-RC, which adapts the recurrent matrix through a low-rank correction driven by streaming prediction errors. The base reservoir and adaptation bases are fixed offline; a small core matrix is adapted online, projected onto a spectral-norm ball, and low-pass filtered at each step. The projection guarantees that every applied recurrent matrix remains within a certified contraction set, and an incremental input-to-state stability bound is established for the reservoir along each online adaptation path, with path-independent rate and gain. On a Lorenz system with an abrupt parameter drift, LoRA-RC cuts post-drift prediction error by 56% versus a fixed RC and 51% versus readout-only adaptation; ablations over 20 seeds show that removing the projection inflates this error by more than a factor of 40.
ER -