Wu Yue, Su Tianhao, Du Bingsheng, Hu Shunbo, Xiong Jie, Pan Deng
Materials Genome Institute, Shanghai University, 200444 Shanghai, China.
Yunnan Province Crystalline Silicon Material Technology Innovation Center, Yunnan Tongwei High Purity Crystalline Silicon Co., Ltd., Baoshan, Yunnan 678000, China.
J Phys Chem Lett. 2024 Dec 19;15(50):12393-12400. doi: 10.1021/acs.jpclett.4c02589. Epub 2024 Dec 10.
In recent years, contrastive learning has gained widespread adoption in machine learning applications to physical systems primarily due to its distinctive cross-modal capabilities and scalability. Building on the foundation of Kolmogorov-Arnold Networks (KANs) [Liu, Z. et al. Kan: Kolmogorov-arnold networks. 2024, 2404.19756], we introduce a novel contrastive learning framework, Kolmogorov-Arnold Contrastive Crystal Property Pretraining (KCCP), which integrates the principles of CLIP and KAN to establish robust correlations between crystal structures and their physical properties. During the training process, we conducted a comparative analysis between Multilayer Perceptron (MLP) and KAN, revealing that KAN significantly outperforms MLP in both accuracy and convergence speed for this task. By extending the capabilities of contrastive learning to the realm of physical systems, KCCP offers a promising approach for constructing cross-data structural and cross-modal physical models, representing an area of considerable potential.
近年来，对比学习因其独特的跨模态能力和可扩展性，在机器学习应用于物理系统方面得到了广泛应用。基于柯尔莫哥洛夫 - 阿诺德网络（KANs）[刘，Z. 等人。KAN：柯尔莫哥洛夫 - 阿诺德网络。2024，2404.19756] 的基础，我们引入了一种新颖的对比学习框架，即柯尔莫哥洛夫 - 阿诺德对比晶体性质预训练（KCCP），它整合了CLIP和KAN的原理，以在晶体结构与其物理性质之间建立稳健的关联。在训练过程中，我们对多层感知器（MLP）和KAN进行了对比分析，结果表明，在该任务中，KAN在准确性和收敛速度方面均显著优于MLP。通过将对比学习的能力扩展到物理系统领域，KCCP为构建跨数据结构和跨模态物理模型提供了一种有前景的方法，代表了一个具有相当潜力的领域。
