Catastrophic forgetting describes the fact that machine learning models will likely forget the knowledge of previously learned tasks after the learning process of a new one. It is a vital problem in the continual learning scenario and recently has attracted tremendous concern across different communities. We explore the catastrophic forgetting phenomena in the context of quantum machine learning. It is found that, similar to those classical learning models based on neural networks, quantum learning systems likewise suffer from such forgetting problem in classification tasks emerging from various application scenes. We show that based on the local geometrical information in the loss function landscape of the trained model, a uniform strategy can be adapted to overcome the forgetting problem in the incremental learning setting. Our results uncover the catastrophic forgetting phenomena in quantum machine learning and offer a practical method to overcome this problem, which opens a new avenue for exploring potential quantum advantages towards continual learning.
Catastrophic forgetting describes the fact that machine learning models will likely forget the knowledge of previously learned tasks after the learning process of a new one. It is a vital problem in the continual learning scenario and recently has attracted tremendous concern across different communities. We explore the catastrophic forgetting phenomena in the context of quantum machine learning. It is found that, similar to those classical learning models based on neural networks, quantum learning systems likewise suffer from such forgetting problem in classification tasks emerging from various application scenes. We show that based on the local geometrical information in the loss function landscape of the trained model, a uniform strategy can be adapted to overcome the forgetting problem in the incremental learning setting. Our results uncover the catastrophic forgetting phenomena in quantum machine learning and offer a practical method to overcome this problem, which opens a new avenue for exploring potential quantum advantages towards continual learning.
Catastrophic forgetting emerges when a neural network’s parameters undergo continuous updates during the sequential training of multiple tasks. The ongoing adaptation, while enhancing the model’s suitability for new tasks, inadvertently leads to a degradation in performance on previously learned tasks. This challenge significantly impedes the sequential learning capabilities essential for the advancement of artificial general intelligence. The phenomenon of catastrophic forgetting also occurs in the field of quantum machine learning, where parametric quantum circuits serve a role analogous to neural networks. In this study, our focus is to explore strategies for mitigating catastrophic forgetting within quantum learning models for quantum data. In this context, we employ a task-based hard attention mechanism, which automatically generates masks for each task to regulate the network’s learning process and resist catastrophic forgetting. To our knowledge, the proposed method is the first one in the field of quantum machine learning to adjust the model structure to prevent catastrophic forgetting. Numerical simulation results demonstrate that our approach preserves the model’s performance on previous tasks without compromising its ability to acquire new knowledge. This effective resistance to catastrophic forgetting marks a significant stride in advancing quantum continual learning.
This paper proposes a novel continual learning model leveraging the unique properties of quantum entanglement to mitigate the catastrophic forgetting problem prevalent in classical machine learning. The core idea involves constructing a "quantum memory" based on entangled states, enabling the encoding, updating, and retrieval of knowledge through controlled manipulation of these states and subsequent quantum measurements. We present a theoretical framework outlining this approach, focusing on the mathematical representation of the entanglement-based learning process. Specifically, we define the state vectors (s1, s2, ..., sK) representing the entangled quantum memory, where K denotes the number of learned concepts. The learning algorithm utilizes unitary transformations (Ui) applied to the state vectors, updating the representation of each concept. Quantum measurement (Mi) is employed to extract information and assess the accuracy of the learned concepts. The central challenge lies in designing the appropriate unitary transformations and measurement operators to effectively capture new information while preserving previously learned knowledge. We explore the mathematical relationships between the state vectors, unitary transformations, and measurement operators, demonstrating how these elements contribute to the continual learning process. This research contributes to the emerging field of quantum machine learning, offering a potential pathway to overcoming the limitations of traditional models in scenarios requiring long-term memory and adaptive learning. The proposed model's ability to maintain and update knowledge without forgetting represents a significant advancement in the field.
Artificial intelligence in dynamic, real-world environments requires the capacity for continual learning. However, standard deep learning suffers from a fundamental issue: loss of plasticity, in which networks gradually lose their ability to learn from new data. Here we show that quantum learning models naturally overcome this limitation, preserving plasticity over long timescales. We demonstrate this advantage systematically across a broad spectrum of tasks from multiple learning paradigms, including supervised learning and reinforcement learning, and diverse data modalities, from classical high-dimensional images to quantum-native datasets. Although classical models exhibit performance degradation correlated with unbounded weight and gradient growth, quantum neural networks maintain consistent learning capabilities regardless of the data or task. We identify the origin of the advantage as the intrinsic physical constraints of quantum models. Unlike classical networks where unbounded weight growth leads to landscape ruggedness or saturation, the unitary constraints confine the optimization to a compact manifold. Our results suggest that the utility of quantum computing in machine learning extends beyond potential speedups, offering a robust pathway for building adaptive artificial intelligence and lifelong learners.
Quantum computers may outperform classical computers on machine learning tasks. Yet, quantum learning systems may suffer from catastrophic forgetting, which is widely believed to be an obstacle to achieving continual learning. Here, we report an experimental demonstration of quantum continual learning on a superconducting processor. In particular, we sequentially train a quantum classifier with three tasks, two about identifying real-life images and one on classifying quantum states, and demonstrate its catastrophic forgetting. To overcome this dilemma, we exploit the elastic weight consolidation strategy and show that the quantum classifier can incrementally retain knowledge across three tasks with an average accuracy exceeding 92.3%. Additionally, for sequential tasks involving quantum-engineered data, we demonstrate that the quantum classifier outperforms a classical classifier with a comparable number of parameters. Our results establish a viable strategy for empowering quantum learning systems with adaptability to sequential tasks.
Artificial intelligence in dynamic, real-world environments requires the capacity for continual learning. However, standard deep learning suffers from a fundamental issue: loss of plasticity, in which networks gradually lose their ability to learn from new data. Here we show that quantum learning models naturally overcome this limitation, preserving plasticity over long timescales. We demonstrate this advantage systematically across a broad spectrum of tasks from multiple learning paradigms, including supervised learning and reinforcement learning, and diverse data modalities, from classical high-dimensional images to quantum-native datasets. Although classical models exhibit performance degradation correlated with unbounded weight and gradient growth, quantum neural networks maintain consistent learning capabilities regardless of the data or task. We identify the origin of the advantage as the intrinsic physical constraints of quantum models. Unlike classical networks where unbounded weight growth leads to landscape ruggedness or saturation, the unitary constraints confine the optimization to a compact manifold. Our results suggest that the utility of quantum computing in machine learning extends beyond potential speedups, offering a robust pathway for building adaptive artificial intelligence and lifelong learners.
Quantum continual learning (QCL) has recently emerged as a promising paradigm, enabling quantum models to incrementally acquire knowledge across tasks while mitigating catastrophic forgetting. Yet, existing approaches face two persistent obstacles: (i) quantum architectures scale poorly to datasets of practical size, and (ii) continuous learning strategies are often entangled with specific quantum models, reducing their adaptability. To address these challenges, we introduce a twofold design. First, we propose hybrid quantum-classical (HQC) networks tailored to different data regimes: HQC1, optimized for Split-MNIST, achieves a133.5%average accuracy improvement over pure quantum baselines, while HQC2 is the first to demonstrate QCL on CIFAR-100, boosting accuracy by 14.66% over a comparable ResNet-18. Second, we present a decoupled QCL framework that separates model architecture from learning strategy, enabling flexible alignment and yielding more than double the average accuracy relative to existing pairings. Together, these contributions establish a scalable and adaptable foundation for practical quantum continuous learning on large, complex datasets.