10 papers
Low-cost passive single-shot ultrafast imaging at 685 Gfps
Dilem EÅlik, Bahadır Utku Kesgin, UÄur TeÄin
Capturing ultrafast transient phenomena conventionally requires streak cameras or computational imaging based on compressed sensing, which lead to complex and costly systems. In th…
Scalable Photonic Neural Networks via Surrogate Scattering-Matrix Inverse Design
Azka Maula Iskandar Muda, UÄur TeÄin
Inverse-designed nanophotonic media are a promising platform for compact optical neural networks, but training them end to end is expensive because each adjoint iteration couples t…
Resolution-Agnostic Lensless Imaging via Fourier Neural Operators
Kerem Ekec, UÄur TeÄin
Lensless cameras based on thin diffusers offer a compact alternative to conventional refractive imaging but rely on computational reconstruction, since the diffuser's point spread…
Recurrent neural networks implemented through spatiotemporal light propagation in optical fibers
Dilem EÅlik, Bahadır Utku Kesgin, UÄur TeÄin
Recurrent neural networks excel at temporal tasks and video processing but require energy-intensive sequential memory operations. We demonstrate that multimode optical fibers natur…
Multimode fiber laser cavities as nonlinear optical processors
Dilem EÅlik, Bahadır Utku Kesgin, Fatma Nur Kılınç +1
Optical computing provides a promising path toward energy-efficient machine learning, yet implementing nonlinear transformations without complex electronics or high-power sources r…
Self-optimizing multichannel optical computing
Fatma Nur Kılınç, UÄur TeÄin
Optical computing offers ultrafast, energy-efficient alternatives to conventional digital processors, yet most implementations remain confined to single-channel processing, severel…