We use Bayesian optimization with a new tunable acquisition function to design a photonic Y-splitter robust to fabrication *** to conventional acquisition functions, our method yields more robust solutions across vary...
详细信息
We report the first proof-of-concept demonstration of fully 3D-printed, transistor-like, no-moving-parts switches capable of performing logic operations. The devices are monolithically made via material extrusion usin...
详细信息
We report the design, fabrication, and characterization of the first monolithically 3D-printed, three-dimensional, three-material, cored inductors for use in compact electromagnetic systems. Additive manufacturing tec...
详细信息
Gallium nitride (GaN)-based high electron mobility transistors (HEMTs) currently define the state-of-the-art in high power density, radiofrequency (RF) semiconductor devices. However, there remain many relatively unex...
详细信息
GaN complementary transistors (CT) are highly desirable for GaN integrated circuits with low static power dissipation [1]. While recent experiments studied the feasibility of GaN CT [2 , 3] and improved GaN p-channel ...
详细信息
There is a growing demand for real-Time image denoising in low-light shooting with ultra-high definition cameras. This paper presents a denoising method that incorporates Haar-wavelet shrinkage denoising and a minimum...
详细信息
This work reports the design, fabrication, and characterization of the first monolithically 3D-printed, three-dimensional inductors for use in compact systems. The novel, air-core, 3D-printed inductors are made via ex...
详细信息
It is fascinating how our body receives an immense amount of sensory information through numerous receptors distributed throughout the body and efficiently integrates it to make decisions for daily activities, while m...
详细信息
It is fascinating how our body receives an immense amount of sensory information through numerous receptors distributed throughout the body and efficiently integrates it to make decisions for daily activities, while maintaining extremely low energy consumption and cognitive load. This bioinspired sensory information processing paradigm offers unparalleled advantages over traditional von Neumann architectures due to its exceptional energy efficiency, fault tolerance, and adaptability. While previous efforts in this area have mainly focused on the development of biomimetic sensors, it is equally important to create a computing architecture that can process sensory data locally before transmitting it to a higher level. The strategy of offloading computation at the edge can significantly reduce data latency, saving transmission bandwidth and relieving the burden of computation at a higher level, just like the way our peripheral nervous system complements the central nervous system.
Linear light-processing functions(e.g.,routing,splitting,filtering)are key functions requiring configuration to implement on a programmable photonic integrated circuit(PPIC).In recirculating waveguide meshes(which inc...
详细信息
Linear light-processing functions(e.g.,routing,splitting,filtering)are key functions requiring configuration to implement on a programmable photonic integrated circuit(PPIC).In recirculating waveguide meshes(which include loop-backs),this is usually done *** previous results describe explorations to perform this task automatically,but their efficiency or applicability is still *** this paper,we propose an efficient method that can automatically realize configurations for many light-processing functions on a square-mesh *** its heart is an automatic differentiation subroutine built upon analytical expressions of scattering matrices that enables gradient descent optimization for functional circuit *** to the state-of-the-art synthesis techniques,our method can realize configurations for a wide range of light-processing functions,and multiple functions on the same PPIC ***,we do not need to separate the functions spatially into different subdomains of the mesh,and the resulting optimum can have multiple functions using the same part of the ***,compared to nongradient-or numerical differentiation-based methods,our proposed approach achieves 3×time reduction in computational cost.
The main purpose of multimodal machine translation (MMT) is to improve the quality of translation results by taking the corresponding visual context as an additional input. Recently many studies in neural machine tran...
详细信息
暂无评论