Search papers, labs, and topics across Lattice.
This paper introduces a quantum incremental learning framework that utilizes trainable mixed-state prototypes to enable the sequential learning of new classes without catastrophic forgetting, addressing the limitations of traditional quantum classifiers in the Noisy Intermediate-Scale Quantum (NISQ) era. By incorporating new class prototypes instead of expanding circuit width, the model maintains efficiency under parameter and memory constraints. The results indicate that this approach achieves effective high-dimensional feature concentration with fewer qubits and reduced computational complexity compared to classical methods, enhancing representation capabilities in incremental learning tasks.
Mixed-state prototypes allow quantum models to learn new classes without expanding circuit complexity, achieving robust performance with fewer qubits.
Incremental learning models are required to learn new classes sequentially without catastrophic forgetting, while operating under parameter and memory constraints. In the Noisy Intermediate-Scale Quantum (NISQ) era, although quantum neural networks offer advantages in feature mapping, hardware limitations restrict circuit width. Furthermore, traditional quantum classifiers are constrained by the number of orthogonal basis states, limiting their capacity to accommodate a continually growing number of categories. Thus, we introduce a novel quantum incremental learning framework based on trainable mixed-state prototypes. Its original design incorporates new classes by adding class prototypes rather than increasing the circuit width of the shared quantum backbone. The use of mixed-state prototypes is another key contribution, since they have representation capabilities to represent information than a single pure-state prototype. And the decomposable mixed-state calculation provides lower production costs and a convenient Hilbert-Schmidt (HS) distance metric for classification. Simulation results show that our model achieves high-dimensional feature concentration using a minimal number of qubits, while demonstrating lower computational complexity and robust representation in incremental learning tasks compared with classical baselines.
