DNNARA: A Deep Neural Network Accelerator using Residue Arithmetic and Integrated Photonics

Abstract

Deep Neural Networks (DNNs) are currently used in many fields, including critical real-time applications. Due to its compute-intensive nature, speeding up DNNs has become an important topic in current research. We propose a hybrid opto-electronic computing architecture targeting the acceleration of DNNs based on the residue number system (RNS). In this novel architecture, we combine the use of Wavelength Division Multiplexing (WDM) and RNS for efficient execution. WDM is used to enable a high level of parallelism while reducing the number of optical components needed to decrease the area of the accelerator. Moreover, RNS is used to generate optical components with short optical critical paths. In addition to speed, this has the advantage of lowering the optical losses and reducing the need for high laser power. Our RNS compute modules use one-hot encoding and thus enable fast switching between the electrical and optical domains. In this work, we demonstrate how to implement the different DNN computational kernels using WDM-enabled RNS based integrated photonics. We provide an accelerator architecture that uses our designed components and perform design space exploration to select efficient architecture parameters. Compared to memristor crossbars, our residue matrix-vector multiplication unit has two orders of magnitude higher peak performance. Our experimental evaluation using DNN benchmarks illustrates that our architecture can perform more than 19 times faster than the state of the art GPUs under the same power budget. CCS CONCEPTS • Computer systems organization → Neural networks; Optical computing; • Computing methodologies → Parallel computing methodologies.

Deep Learning Image Analysis of Nanoplasmonic Sensors: Toward Medical Breath Monitoring

“…To thoroughly study the correlations between the sensor responses (RGB values) with various concentrations of acetone and other interfering analytes, a deep learning neural network model named SensingNet was developed in this work to intelligently analyze the change of RGB values in each sensor image and accurately predict the NPP sensing results. Deep learning neural networks have been developed and applied to address challenges in many areas. − …”

Quantifying Information via Shannon Entropy in Spatially Structured Optical Beams

“…The constraints of probability theory on the definition of information and of quantum mechanics on conjugate observables are satisfied working around the properties of the WDF-a pseudo-probability distribution. We foresee the relevance of this formalism in the context of recent developments for (i) free-space information-processing optics [34]; (ii) integrated photonics-based information processing [35] such as neural network-based accelerators [36] and photonic tensor cores [37]; (iii) adaptive sensing [38]; and (iv) analog optical and photonic processors [39][40][41]. As the data compression coefficient is naturally bounded by Shannon information, carried by the beam [42], this work indirectly points towards higher information capacity in beams with a nontrivial structure, like HG, LG, and Bessel-Gauss modes [43,44].…”

Cryptographic Circuit Technology Consisting of Optical Logic Gates

“…Optical circuits consisting of optical logic gates to enable logical operations constitute one example of optical technology. It has been shown, for example, that optical circuits can be used for the computational operations required by learning algorithms in the field of deep learning and that lowlatency and low-power operations can be achieved [1].…”