Abstract:Electromagnetic field simulation is central to designing, optimizing, and validating photonic devices and circuits. However, costly computation associated with numerical simulation poses a significant bottleneck, hindering scalability and turnaround time in the photonic circuit design process. Neural operators offer a promising alternative, but existing SOTA approaches, NeurOLight, struggle with predicting high-fidelity fields for real-world complicated photonic devices, with the best reported 0.38 normalized mean absolute error in NeurOLight. The inter-plays of highly complex light-matter interaction, e.g., scattering and resonance, sensitivity to local structure details, non-uniform learning complexity for full-domain simulation, and rich frequency information, contribute to the failure of existing neural PDE solvers. In this work, we boost the prediction fidelity to an unprecedented level for simulating complex photonic devices with a novel operator design driven by the above challenges. We propose a novel cross-axis factorized PACE operator with a strong long-distance modeling capacity to connect the full-domain complex field pattern with local device structures. Inspired by human learning, we further divide and conquer the simulation task for extremely hard cases into two progressively easy tasks, with a first-stage model learning an initial solution refined by a second model. On various complicated photonic device benchmarks, we demonstrate one sole PACE model is capable of achieving 73% lower error with 50% fewer parameters compared with various recent ML for PDE solvers. The two-stage setup further advances high-fidelity simulation for even more intricate cases. In terms of runtime, PACE demonstrates 154-577x and 11.8-12x simulation speedup over numerical solver using scipy or highly-optimized pardiso solver, respectively. We open sourced the code and dataset.

PROTEUS: Rule-Based Self-Adaptation in Photonic NoCs for Loss-Aware Co-Management of Laser Power and Performance

Managing Laser Power in Silicon-Photonic NoC Through Cache and NoC Reconfiguration

A novel cross-layer framework for early-stage power delivery and architecture co-exploration.

DLPS: Dynamic Laser Power Scaling for Optical Network-on-Chip

Energy-Efficient Hybrid Optical-Electronic Network-On-Chip for Future Many-Core Processors

ARXON: A Framework for Approximate Communication over Photonic Networks-on-Chip

Energy-Proportional Data Center Network Architecture Through OS, Switch and Laser Co-design

Proactive flow control using adaptive beam forming for smart intra-layer data communication in wireless network on chip

Modeling and Performance Analysis of a Fault-Tolerant 3D Photonic Network-on-Chip Based on Hybrid Photonics-Plasmonics

Managing power consumption in networks on chips

Poet: A Power Efficient Hybrid Optical NoC Topology for Heterogeneous CPU-GPU Systems

Exploring a New Adaptive Routing Based on the Dijkstra Algorithm in Optical Networks-on-Chip

A High-Performance Low-Power Nanophotonic On-Chip Network

Efficient On-Chip Learning for Optical Neural Networks Through Power-Aware Sparse Zeroth-Order Optimization

PACE: Pacing Operator Learning to Accurate Optical Field Simulation for Complicated Photonic Devices

Exploration of Power-Savings on Multi-Core Architectures With Offloaded Real-Time Operating System

Design Technology for Scalable and Robust Photonic Integrated Circuits: Invited Paper

Software-defined optical networking applications enabled by programmable integrated photonics

Re-architecting Congestion Management in Lossless Ethernet.

ADEPT-Z: Zero-Shot Automated Circuit Topology Search for Pareto-Optimal Photonic Tensor Cores

Optical Versus Electrical: Performance Evaluation of Network On-Chip Topologies for UWASN Manycore Processors