Curvature-aware dynamic precision approach for physics-informed neural networks

Physics-informed neural networks (PINNs) have become a promising framework for simulating partial differential equations (PDEs) by embedding physical laws directly into neural network training. However, recent studies show that PINN optimisation is sensitive to numerical precision. Existing implementations commonly use either single precision (FP32), which is computationally efficient but prone to failure modes, or double precision (FP64), which is robust but substantially expensive. This creates a trade-off between computational efficiency and numerical accuracy. To reduce the computational cost of double-precision training while retaining prediction accuracy, we propose a curvature-aware precision controller that adapts numerical precision during training rather than treating it as a fixed implementation choice. The proposed method reuses curvature information derived from the limited-memory BFGS (L-BFGS) optimiser to construct a precision controller, retaining FP32 when lower precision is sufficient and promoting computation to FP64 when the training dynamics indicate numerical sensitivity or precision-limited stagnation. We evaluate the proposed approach on four canonical PINN failure-mode benchmarks and an irradiance-driven ordinary differential equation example. We further test the proposed approach across different neural network architectures. The method consistently matches or even slightly exceeds full FP64 solution accuracy while reducing training time relative to full double-precision training on all benchmark equations. The obtained results indicate that precision sensitivity in PINN optimisation is phase-dependent, and that selectively applying higher precision only during numerically critical stages can lower computational cost without sacrificing predictive accuracy.

Real-Time Wave Mitigation for Water-Air OWC Systems Via Beam Tracking

In a water-air optical wireless communication (OWC) channel, dynamic ocean waves may significantly deflect the light beam from its original direction, thus deteriorating the communication performance. In this letter, a beam tracking system to mitigate wave-induced communication degradation is experimentally demonstrated. By employing the beam tracking on the PAM6 system, a maximum of 486% improvement (from 140 Mb/s to 820 Mb/s) in throughput is realized at a 1.2-m air distance for an average wave slope changing rate of 0.34 rad/s. For PAM4 signals, the packet loss rate reduces significantly from 75% to 11%, and a maximum throughput of 1.25 Gb/s is achieved.