相关论文: Energy-efficient neural network inference with mic…

Optical neural networks (ONNs) have demonstrated record-breaking potential in high-performance neuromorphic computing due to their ultra-high execution speed and low energy consumption. However, current learning protocols fail to provide…

Plastic self-adaptation, nonlinear recurrent dynamics and multi-scale memory are desired features in hardware implementations of neural networks, because they enable them to learn, adapt and process information similarly to the way…

Exciton-polaritons offer a versatile platform for realization of all-optical integrated logic gates due to the strong effective optical nonlinearity resulting from the exciton-exciton interactions. In most of the current excitonic materials…

Neural networks are one of the first major milestones in developing artificial intelligence systems. The utilisation of integrated photonics in neural networks offers a promising alternative approach to microelectronic and hybrid…

The formation of exciton-polaritons allows the transport of energy over hundreds of nanometres at velocities up to 10^6 m s^-1 in organic semiconductors films in the absence of external cavity structures.

We introduce a novel photonic neural network using photonic crystal fibers, leveraging femtosecond pulse supercontinuum generation for optical computing. Investigating its efficacy across machine learning tasks, we uncover the crucial…

We present a fully microscopic approach to the transition rate of two exciton-photon polaritons. The non-trivial consequences of the polariton composite nature -- here treated exactly through a development of our composite-exciton many-body…

The recent surge of interest in Deep Neural Networks (DNNs) has led to increasingly complex networks that tax computational and memory resources. Many DNNs presently use 16-bit or 32-bit floating point operations. Significant performance…

We consider non-dissipative drag between Bose-condensed exciton polaritons in optical microcavity and embedded superconductors. This effect consists in induction of a non-dissipative electric current in the superconductor by motion of…

Research has shown that deep neural networks contain significant redundancy, and that high classification accuracies can be achieved even when weights and activations are quantised down to binary values. Network binarisation on FPGAs…

Neuromorphic computing, inspired by the brain, promises extreme efficiency for certain classes of learning tasks, such as classification and pattern recognition. The performance and power consumption of neuromorphic computing depends…

Exciton-polaritons are hybrid elementary excitations of light and matter that, thanks to their nonlinear properties, enable a plethora of physical phenomena ranging from room temperature condensation to superfluidity. While polaritons are…

One of the most computationally intensive parts in modern recognition systems is an inference of deep neural networks that are used for image classification, segmentation, enhancement, and recognition. The growing popularity of edge…

Efficient machine learning inference is essential for the rapid adoption of artificial intelligence across various domains.On-chip optical computing has emerged as a transformative solution for accelerating machine learning tasks, owing to…

Exciton-polaritons in semiconductor microcavities generate fascinating effects such as long-range spatial coherence and Bose-Einstein Condensation (BEC), which are attractive for their potential use in low threshold lasers, vortices and…

Exciton-polaritons, hybrid photon-exciton quasiparticles, constitute a useful platform for the study of light-matter interaction and nonlinear photonic applications. In this work, we realize a monolithic Tamm-plasmon microcavity embedding a…

Improving the efficiency of neural network inference is undeniably important in a time where commercial use of AI models increases daily. Node pruning is the art of removing computational units such as neurons, filters, attention heads, or…

Designing energy-efficient networks is of critical importance for enabling state-of-the-art deep learning in mobile and edge settings where the computation and energy budgets are highly limited. Recently, Liu et al. (2019) framed the search…

Large-scale artificial neural networks have shown significant promise in addressing a wide range of classification and recognition applications. However, their large computational requirements stretch the capabilities of computing…

Optical neural networks promise ultrafast, low-energy information processing by performing computation directly with photons. Current implementations, however, are largely restricted to steady-state operation and rely on high-latency…