8 papers · 1 filter
Physical policy gradient theorem for in situ stochastic-adjoint training
William Tuxbury, Zin Lin
In situ adjoint training extracts parameter gradients directly from measurement, but has so far been limited to reciprocal or restricted systems. Here, we introduce the physical co…
In-situ Adjoint Wave Control in Reconfigurable Non-Hermitian Nonlinear Systems
John Guillamon, William Tuxbury, Cheng-Zhen Wang +3
Complex multipath environments are usually avoided in wave-based information processing because repeated scattering creates many interfering propagation paths, obscuring controllab…
Differentiable Fast Far-Field Transform in Cylindrical Coordinates for Large-Area Cascaded Metalens Optics
Arvin Keshvari, Ata Shakeri, William Tuxbury +4
We present a fully differentiable far-field transform in cylindrical coordinates for full-area point spread function (PSF) evaluation and optimization of large axisymmetric metalen…
Adaptive Sensing beyond Non-Adaptive Information Limits: End-to-End Co-Design of Geometry, Policy, and Inference
Arvin Keshvari, William Tuxbury, Zin Lin
Inverse design has transformed vast physical parameter spaces into a substrate for emergent functionality, raising the tantalizing prospect of relocating intelligence from the digi…
Synthetic Reflectionless Mode Exceptional Degeneracies via Emergent Local Symmetries
William Tuxbury, Lucas Fernandez-Alcazar, Tsampikos Kottos
We propose a new kind of physically realizable exceptional point degeneracies (EPDs) corresponding to synthetic reflectionless modes (SRM). These are solutions of an auxiliary wave…
Unidirectional Amplification in the Frozen Mode Regime Enabled by a Nonlinear Defect
Serena Landers, William Tuxbury, Ilya Vitebskiy +1
A stationary inflection point (SIP) is a spectral singularity of the Bloch dispersion relation of a periodic structure where the first and the second derivatives of with…