Deep Learning with Passive Optical Nonlinear Mapping

• URL: http://arxiv.org/abs/2307.08558v2
• Date: Tue, 18 Jul 2023 15:23:48 GMT
• Title: Deep Learning with Passive Optical Nonlinear Mapping
• Authors: Fei Xia, Kyungduk Kim, Yaniv Eliezer, Liam Shaughnessy, Sylvain Gigan, Hui Cao
• Abstract summary: We introduce a design that leverages multiple scattering in a reverberating cavity to passively induce optical nonlinear random mapping. We show we can perform optical data compression, facilitated by multiple scattering in the cavity, to efficiently compress and retain vital information. Our findings pave the way for novel algorithms and architectural designs for optical computing.
• Score: 9.177212626554505
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Deep learning has fundamentally transformed artificial intelligence, but the ever-increasing complexity in deep learning models calls for specialized hardware accelerators. Optical accelerators can potentially offer enhanced performance, scalability, and energy efficiency. However, achieving nonlinear mapping, a critical component of neural networks, remains challenging optically. Here, we introduce a design that leverages multiple scattering in a reverberating cavity to passively induce optical nonlinear random mapping, without the need for additional laser power. A key advantage emerging from our work is that we show we can perform optical data compression, facilitated by multiple scattering in the cavity, to efficiently compress and retain vital information while also decreasing data dimensionality. This allows rapid optical information processing and generation of low dimensional mixtures of highly nonlinear features. These are particularly useful for applications demanding high-speed analysis and responses such as in edge computing devices. Utilizing rapid optical information processing capabilities, our optical platforms could potentially offer more efficient and real-time processing solutions for a broad range of applications. We demonstrate the efficacy of our design in improving computational performance across tasks, including classification, image reconstruction, key-point detection, and object detection, all achieved through optical data compression combined with a digital decoder. Notably, we observed high performance, at an extreme compression ratio, for real-time pedestrian detection. Our findings pave the way for novel algorithms and architectural designs for optical computing.

Related papers

• Successive optimization of optics and post-processing with differentiable coherent PSF operator and field information [9.527960631238173]
  We introduce a precise optical simulation model, and every operation in pipeline is differentiable. To efficiently address various degradation, we design a joint optimization procedure that leverages field information.
  arXiv  Detail & Related papers  (2024-12-19T07:49:40Z)
• Generalizable Non-Line-of-Sight Imaging with Learnable Physical Priors [52.195637608631955]
  Non-line-of-sight (NLOS) imaging has attracted increasing attention due to its potential applications. Existing NLOS reconstruction approaches are constrained by the reliance on empirical physical priors. We introduce a novel learning-based solution, comprising two key designs: Learnable Path Compensation (LPC) and Adaptive Phasor Field (APF)
  arXiv  Detail & Related papers  (2024-09-21T04:39:45Z)
• Complex-valued universal linear transformations and image encryption using spatially incoherent diffractive networks [0.0]
  As an optical processor, a Diffractive Deep Neural Network (D2NN) utilizes engineered diffractive surfaces designed through machine learning to perform all-optical information processing. We show that a spatially incoherent diffractive visual processor can approximate any complex-valued linear transformation and be used for all-optical image encryption using incoherent illumination.
  arXiv  Detail & Related papers  (2023-10-05T08:43:59Z)
• Nonlinear Processing with Linear Optics [3.3998740964877467]
  We present a novel framework that uses multiple scattering that is capable of synthesizing programmable linear and nonlinear transformations concurrently at low optical power. We empirically found that scaling of this optical framework follows the power law as in state-of-the-art deep digital networks.
  arXiv  Detail & Related papers  (2023-07-17T14:49:06Z)
• Characterization of multi-mode linear optical networks [0.0]
  We formulate efficient procedures for the characterization of optical circuits in the presence of imperfections. We show the viability of this approach in an experimentally relevant scenario, defined by a tunable integrated photonic circuit. Our findings can find application in a wide range of optical setups, based both on bulk and integrated configurations.
  arXiv  Detail & Related papers  (2023-04-13T13:09:14Z)
• On-chip quantum information processing with distinguishable photons [55.41644538483948]
  Multi-photon interference is at the heart of photonic quantum technologies. Here, we experimentally demonstrate that detection can be implemented with a temporal resolution sufficient to interfere photons detuned on the scales necessary for cavity-based integrated photon sources. We show how time-resolved detection of non-ideal photons can be used to improve the fidelity of an entangling operation and to mitigate the reduction of computational complexity in boson sampling experiments.
  arXiv  Detail & Related papers  (2022-10-14T18:16:49Z)
• All-Photonic Artificial Neural Network Processor Via Non-linear Optics [0.0]
  We propose an all-photonic artificial neural network processor. Information is encoded in the amplitudes of frequency modes that act as neurons. Our architecture is unique in providing a completely unitary, reversible mode of computation.
  arXiv  Detail & Related papers  (2022-05-17T19:55:30Z)
• All-optical graph representation learning using integrated diffractive photonic computing units [51.15389025760809]
  Photonic neural networks perform brain-inspired computations using photons instead of electrons. We propose an all-optical graph representation learning architecture, termed diffractive graph neural network (DGNN) We demonstrate the use of DGNN extracted features for node and graph-level classification tasks with benchmark databases and achieve superior performance.
  arXiv  Detail & Related papers  (2022-04-23T02:29:48Z)
• Topologically Protecting Squeezed Light on a Photonic Chip [58.71663911863411]
  Integrated photonics offers an elegant way to increase the nonlinearity by confining light strictly inside the waveguide. We experimentally demonstrate the topologically protected nonlinear process of spontaneous four-wave mixing enabling the generation of squeezed light on a silica chip.
  arXiv  Detail & Related papers  (2021-06-14T13:39:46Z)
• Learning optical flow from still images [53.295332513139925]
  We introduce a framework to generate accurate ground-truth optical flow annotations quickly and in large amounts from any readily available single real picture. We virtually move the camera in the reconstructed environment with known motion vectors and rotation angles. When trained with our data, state-of-the-art optical flow networks achieve superior generalization to unseen real data.
  arXiv  Detail & Related papers  (2021-04-08T17:59:58Z)
• Rapid characterisation of linear-optical networks via PhaseLift [51.03305009278831]
  Integrated photonics offers great phase-stability and can rely on the large scale manufacturability provided by the semiconductor industry. New devices, based on such optical circuits, hold the promise of faster and energy-efficient computations in machine learning applications. We present a novel technique to reconstruct the transfer matrix of linear optical networks.
  arXiv  Detail & Related papers  (2020-10-01T16:04:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.