Quantum Transfer Learning Shows Improved Robustness in Low-Data Regimes
For practitioners in transfer learning with limited data, this work provides empirical evidence that quantum models may offer robustness advantages over classical models, though the gains are incremental and specific to low-data regimes.
This paper systematically compares quantum and classical models in transfer learning under limited data, finding that quantum models exhibit significantly less accuracy degradation and more stable performance, indicating improved robustness and data efficiency.
Transfer learning under limited data is a challenging setting, where models must adapt to new tasks with minimal supervision. Prior work has primarily focused on improving absolute accuracy in transfer learning. However, empirical evidence comparing quantum and classical models in realistic transfer learning settings remains limited, especially in low-data regimes. In this work, we systematically study the robustness of quantum models under reduced training data. We evaluate multiple quantum and classical architectures across diverse transfer tasks and retraining configurations, and quantify robustness using accuracy degradation and relative performance retention (RPR). Our results show that, although classical models often achieve higher peak performance, they exhibit significantly larger degradation when training data is limited. In contrast, quantum models maintain more stable performance across data regimes, indicating improved robustness and data efficiency. These findings provide empirical evidence that quantum models can offer improved robustness in low-resource transfer learning scenarios.