Makoto Naruse 교수

Reinforcement learning involves decision making in dynamic and uncertain environments and constitutes an important element of artificial intelligence (AI). In this work, we experimentally demonstrate that the ultrafast chaotic oscillatory dynamics of lasers efficiently solve the multi-armed bandit problem (MAB), which requires decision making concerning a class of difficult trade-offs called the exploration-exploitation dilemma. To solve the MAB, a certain degree of randomness is required for ex

Decision making is critical in our daily lives and for society in general and is finding evermore practical applications in information and communication technologies. Herein, we demonstrate experimentally that single photons can be used to make decisions in uncertain, dynamically changing environments. Using a nitrogen-vacancy in a nanodiamond as a single-photon source, we demonstrate the decision-making capability by solving the multi-armed bandit problem. This capability is directly and immed

Reinforcement learning involves decision-making in dynamic and uncertain environments and constitutes a crucial element of artificial intelligence. In our previous work, we experimentally demonstrated that the ultrafast chaotic oscillatory dynamics of lasers can be used to efficiently solve the two-armed bandit problem, which requires decision-making concerning a class of difficult trade-offs called the exploration-exploitation dilemma. However, only two selections were employed in that research

Nanophotonics has been extensively studied with the aim of unveiling and exploiting light-matter interactions that occur at a scale below the diffraction limit of light, and recent progress made in experimental technologies--both in nanomaterial fabrication and characterization--is driving further advancements in the field. From the viewpoint of information, on the other hand, novel architectures, design and analysis principles, and even novel computing paradigms should be considered so that we

Optical near-field interactions exhibit different behavior at different scales, which we term scale-dependent physical hierarchy. Using the intrinsic logical hierarchy of information and a simple digital coding scheme, scale-dependent optical memory accesses are associated with different levels of the information hierarchy. The basic principle is demonstrated by finite-different time-domain simulations and experiments using metal nanoparticles.

SUMMARY We approach nanophotonic computing on the basis of op-tical near-field interactions between quantum dots. A table lookup, ormatrix-vector multiplication, architecture is proposed. As fundamentalfunctionality, a data summation mechanism and digital-to-analog conver-sion are experimentally demonstrated using CuCl quantum dots. Owing tothe diffraction-limit-free nature of nanophotonics, these architectures canachieve ultrahigh density integration compared to conventional bulky opti-cal system

Nature-inspired devices and architectures are attracting considerable attention for various purposes, including developing novel computing based on spatiotemporal dynamics, exploiting stochastic processes for computing, and reducing energy dissipation. This paper demonstrates that the optical energy transfer between quantum nanostructures mediated by optical near-field interactions occurring at scales far below the wavelength of light could be utilized for solving constraint satisfaction problem

A nanoscale data summation architecture is proposed and experimentally demonstrated based on the optical near-field interaction between quantum dots. Based on local electromagnetic interactions between a few nanometric elements via optical near fields, we can combine multiple excitations at a certain quantum dot, which allows construction of a summation architecture. Summation plays a key role for content-addressable memory, which is one of the most important functions in optical networks.

Optical near-fields exhibit different behavior at different scales, a unique feature that could be exploited in system applications. Here we theoretically analyze the hierarchical nature of optical near-field interactions based on the angular spectrum representation, which allows analytical treatment giving an intuitive picture of the localization of optical near-fields and representing the relevance/irrelevance in optical near-field interactions at different scales of observation in the sub-wav

We theoretically analyzed the lower bound of energy dissipation required for optical excitation transfer from smaller quantum dots to larger ones via optical near-field interactions. The coherent interaction between two quantum dots via optical near-fields results in unidirectional excitation transfer by an energy dissipation process occurring in the larger dot. We investigated the lower bound of this energy dissipation, or the intersublevel energy difference at the larger dot, when the excitati

We present tamper resistance in optical excitation transfer via optical near-field interactions based on the energy dissipation process occurring locally in nanometric devices such as quantum dots. A theoretical comparison with electrical systems is also shown, focusing on the required environmental conditions. Numerical simulations based on virtual photon models demonstrate high tamper resistance.

Optics has been playing crucial roles in security applications ranging from authentication and watermarks to anti-counterfeiting. However, since the fundamental physical principle involves optical far-fields, or propagating light, diffraction of light causes severe difficulties, for example in device scaling and system integration. Moreover, conventional security technologies in use today have been facing increasingly stringent demands to safeguard against threats such as counterfeiting of holog

Ultrahigh-density data-broadcasting optical interconnects are proposed and experimentally demonstrated using optical near-field interactions between quantum dots, which cannot be driven by far-field light, allowing sub-wavelength device operation, and far-field excitation for global interconnects. The proposed scheme helps to solve interconnection difficulties experienced in nano-scale device arrays since components for individually guiding light from external systems are not required. Combining

We theoretically and experimentally investigated a system composed of a mixture of different-sized quantum dots involving optical near-field interactions to effectively induce optical excitation transfer. We demonstrated that the ratio of the number of smaller quantum dots to larger ones can be optimized using a density-matrix formalism so that excitons generated in the smaller ones are efficiently transferred to the larger ones. We also describe experimental demonstrations based on a mixture of

Decision making in dynamically changing uncertain environments is one of the most important elements in information and communications technology, in applications ranging from resource assignment in data centers to wireless communications and search functions. Here, we review our research on the efficient physical realization or acceleration of decision making using photonics. The problem of interest is the multi-armed bandit (MAB) problem, where the purpose is to maximize the total rewards in u