Kaiwu-PyTorch-Plugin: Bridging Deep Learning and Photonic Quantum Computing for Energy-Based Models and Active Sample Selection
This work addresses computational bottlenecks in deep learning for researchers working with Energy-Based Models by providing a quantum-classical integration framework.
The paper tackles classical inefficiencies in Energy-Based Models by introducing the Kaiwu-PyTorch-Plugin (KPP), which integrates a Coherent Ising Machine into PyTorch to accelerate Boltzmann sampling and optimize training data via Active Sampling, achieving state-of-the-art performance on single-cell and OpenWebText datasets.
This paper introduces the Kaiwu-PyTorch-Plugin (KPP) to bridge Deep Learning and Photonic Quantum Computing across multiple dimensions. KPP integrates the Coherent Ising Machine into the PyTorch ecosystem, addressing classical inefficiencies in Energy-Based Models. The framework facilitates quantum integration in three key aspects: accelerating Boltzmann sampling, optimizing training data via Active Sampling, and constructing hybrid architectures like QBM-VAE and Q-Diffusion. Empirical results on single-cell and OpenWebText datasets demonstrate KPPs ability to achieve SOTA performance, validating a comprehensive quantum-classical paradigm.