Laser & Photonics Review · 2022 · 56 citations · 44 references
Natural SciencesMaterials EngineeringMaterials ScienceMachine LearningOpticsOptoelectronicsDeep LearningElectro-optic MaterialsNeural NetworksOptical Thin FilmsOptoelectronic DevicesOptoelectronic MaterialsClassification AccuracyClassification Accuracy AssessmentEnergy ConsumptionHandwritten Character RecognitionMetasurfacesNetwork CapacityNeural Network-based RecognitionOptical Computing
Abstract Optical computing has a series of advantages over its electronic counterpart, e.g., low energy consumption, high speed, and intrinsic parallelism. Diffraction deep neural networks (D 2 NNs) are a prominent example capable of processing images directly without addressing the spatial locations of each element. Despite the great successes, the D 2 NNs typically utilize the multilayer framework and face the severe challenge of misalignment in the optical region. Herein, a single metasurface‐based optical‐electronic hybrid neural network (OENN) is proposed and experimentally demonstrated. The OENN is composed of a titanium dioxide (TiO 2 ) metasurface and a fully‐connected electronic layer. The combination of nonlocal neural layer and nonlinear transformation has significantly expanded the neural network capacity. Consequently, the classification accuracy on handwritten digits recognition can still be as high as 98.05% without employing the architecture of cascaded metasurfaces. The OENN shall shed light on the practical applications of optical computing in the visible spectrum.
44
Speech recognition with deep recurrent neural networks
Alex Graves, Abdelrahman Mohamed, Geoffrey E. Hinton · 2013 · 8.8K citations
Approximation capabilities of multilayer feedforward networks
Kurt Hornik · Neural Networks · 1991 · 6.1K citations