Interaction Field Matching: Overcoming Limitations of Electrostatic Models
This work addresses a specific technical bottleneck in physics-inspired data generation methods, offering an incremental improvement over existing approaches.
The paper tackles the limitations of Electrostatic Field Matching (EFM) for data generation and transfer by proposing Interaction Field Matching (IFM), a generalization that uses general interaction fields, and demonstrates its performance on toy and image data transfer problems.
Electrostatic field matching (EFM) has recently appeared as a novel physics-inspired paradigm for data generation and transfer using the idea of an electric capacitor. However, it requires modeling electrostatic fields using neural networks, which is non-trivial because of the necessity to take into account the complex field outside the capacitor plates. In this paper, we propose Interaction Field Matching (IFM), a generalization of EFM which allows using general interaction fields beyond the electrostatic one. Furthermore, inspired by strong interactions between quarks and antiquarks in physics, we design a particular interaction field realization which solves the problems which arise when modeling electrostatic fields in EFM. We show the performance on a series of toy and image data transfer problems.