We show how the array of convolutional adaptive interconnections needed for deep
learning can be physically implemented and learned in an all-optical multistage dynamic holographically-interconnected architecture using thick Fourier-plane dynamic holographic photorefractive crystals.
This optical architecture is self-aligned, phase-calibrated, and aberration compensated by using...
Often, when speaking of artificial intelligence, we think of software systems capable of learning-memorizing-reasoning. All AI codes run on traditional computers, which suffer from architectures not optimized for neural computing: in fact, they are based on traditional von Neumann geometry, for which processing and memory areas are physically separated. Unlike computers, the brain uses a...
The ever-increasing need for processing large volumes of data is currently met with electronic processors. However, their limitations such as high power dissipation necessitate new modalities for computing. Optics is a promising contender for lower power operation. However, to achieve computing with the efficiency of light, creating and controlling nonlinearity optically stands as the next...
The newborn field of gravitational wave astronomy is gaining speed and after the first detection in 2015, gravitational waves coming from almost a hundred compact binary coalescences were detected between 2015 and 2020. Among all GW detections, more than two third were detected in the last observation run (O3). This was possible only thanks to the various upgrades implemented in between...
We show that undoped $LiNbO_3$ films constitute an attractive photosensitive medium to form reconfigurable optical functions controlled by pyroelectric effect. The present work illustrates this potential through the formation light-induced tunable X-junctions capable of fulfilling the function of optical neurons.
The $LiNbO_3$ films is fabricated from a photonic grade $LiNbO_3$ wafer...
Ferroelectric crystals can modify their optical properties by producing photo-excited carriers when exposed to moderate laser power. This change stems from the direction-dependent photovoltages developed along the c-axis of the ferroelectric crystal. Here, we harness the pyroelectric, photorefractive and photoelastic effects to induce reversible (pyroelectric and photoelastic) and irreversible...
The realization of low-loss, high-speed all-optical interconnections is of great interest for possible photonic applications; having plastic interconnections will extend applications to the field of neural networks. To mimic biological neuroplasticity, interconnections should be able to be created, deleted, strengthened, and weakened due to the learning and reasoning purposes of neural...