--It is shown that the port behavior of a resistor-diode network corresponds to the solution of a ReLU monotone operator equilibrium network (a neural network in the limit of infinite depth), giving a parsimonious construction of a neural network in analog hardware. We furthermore show that the gradient of such a circuit can be computed directly in hardware, using a procedure we call hardware linearization . This allows the network to be trained in hardware, which we demonstrate with a device-level circuit simulation. We extend the results to cascades of resistor-diode networks, which can be used to implement feedforward and other asymmetric networks. We finally show that different nonlinear elements give rise to different activation functions, and introduce the novel diode ReLU which is induced by a non-ideal diode model. The idea of building a neural network in analog hardware is classical [1]-[5]. Since the discovery of semiconductor devices with memristive properties [6], and in light of the growing energy intensiveness of machine learning systems, there has been a resurgence of interest in building devices which incorporate analog memristive components and are specially suited for deep learning applications [7], [8]. One of the primary advantages of such devices is that memristors, and similar elements such as phase change memory, act as both memory and computational units. This allows the transport delay between memory and computation to be circumvented. A particularly successful design is to arrange a number of memristors in a crossbar array, which can be used to perform matrix-vector calculation in a single operation [9]-[12].

As artificial intelligence continues to push into real-time, edge-based and resource-constrained environments, there is an urgent need for novel, hardware-efficient computational models. In this study, we present and validate a neuromorphic computing architecture based on resonant-tunnelling diodes (RTDs), which exhibit the nonlinear characteristics ideal for physical reservoir computing (RC). We theoretically formulate and numerically implement an RTD-based RC system and demonstrate its effectiveness on two image recognition benchmarks: handwritten digit classification and object recognition using the Fruit~360 dataset. Our results show that this circuit-level architecture delivers promising performance while adhering to the principles of next-generation RC -- eliminating random connectivity in favour of a deterministic nonlinear transformation of input signals.

Reconfigurable intelligent surface (RIS) is a two-dimensional periodic structure integrated with a large number of reflective elements, which can manipulate electromagnetic waves in a digital way, offering great potentials for wireless communication and radar detection applications. However, conventional RIS designs highly rely on extensive full-wave EM simulations that are extremely time-consuming. To address this challenge, we propose a machine-learning-assisted approach for efficient RIS design. An accurate and fast model to predict the reflection coefficient of RIS element is developed by combining a multi-layer perceptron neural network (MLP) and a dual-port network, which can significantly reduce tedious EM simulations in the network training. A RIS has been practically designed based on the proposed method. To verify the proposed method, the RIS has also been fabricated and measured. The experimental results are in good agreement with the simulation results, which validates the efficacy of the proposed method in RIS design.

Resistor networks have recently had a surge of interest as substrates for energy-efficient self-learning machines. This work studies the computational capabilities of these resistor networks. We show that electrical networks composed of voltage sources, linear resistors, diodes and voltage-controlled voltage sources (VCVS) can implement any continuous functions. To prove it, we assume that the circuit elements are ideal and that the conductances of variable resistors and the amplification factors of the VCVS's can take arbitrary values -- arbitrarily small or arbitrarily large. The constructive nature of our proof could also inform the design of such self-learning electrical networks.

We present a directed percolation inverse problem for diode networks: Given information about which pairs of nodes allow current to percolate from one to the other, can one find a configuration of diodes consistent with the observed currents? We implement a divide-and-concur iterative projection method for solving the problem and demonstrate the supremacy of our method over an exhaustive approach for nontrivial instances of the problem. We find that the problem is most difficult when some but not all of the percolation data are hidden, and that the most difficult networks to reconstruct generally are those for which the currents are most sensitive to the addition or removal of a single diode.

In this paper, two methodologies are used to speed up the maximization of the breakdown volt-age (BV) of a vertical GaN diode that has a theoretical maximum BV of ~2100V. Firstly, we demonstrated a 5X faster accurate simulation method in Technology Computer-Aided-Design (TCAD). This allows us to find 50% more numbers of high BV (>1400V) designs at a given simulation time. Secondly, a machine learning (ML) model is developed using TCAD-generated data and used as a surrogate model for differential evolution optimization. It can inversely design an out-of-the-training-range structure with BV as high as 1887V (89% of the ideal case) compared to ~1100V designed with human domain expertise.

The news: A new type of artificial eye, made by combining light-sensing electronics with a neural network on a single tiny chip, can make sense of what it's seeing in just a few nanoseconds, far faster than existing image sensors. Why it matters: Computer vision is integral to many applications of AI--from driverless cars to industrial robots to smart sensors that act as our eyes in remote locations--and machines have become very good at responding to what they see. But most image recognition needs a lot of computing power to work. Part of the problem is a bottleneck at the heart of traditional sensors, which capture a huge amount of visual data, regardless of whether or not it is useful for classifying an image. Crunching all that data slows things down.

Compact semiconductor device models are essential for efficiently designing and analyzing large circuits. However, traditional compact model development requires a large amount of manual effort and can span many years. Moreover, inclusion of new physics (eg, radiation effects) into an existing compact model is not trivial and may require redevelopment from scratch. Machine Learning (ML) techniques have the potential to automate and significantly speed up the development of compact models. In addition, ML provides a range of modeling options that can be used to develop hierarchies of compact models tailored to specific circuit design stages. In this paper, we explore three such options: (1) table-based interpolation, (2)Generalized Moving Least-Squares, and (3) feed-forward Deep Neural Networks, to develop compact models for a p-n junction diode. We evaluate the performance of these "data-driven" compact models by (1) comparing their voltage-current characteristics against laboratory data, and (2) building a bridge rectifier circuit using these devices, predicting the circuit's behavior using SPICE-like circuit simulations, and then comparing these predictions against laboratory measurements of the same circuit.

Existing VR systems and experiences are immersive, engaging and sometimes even interactive. But they don't offer a quick, easy way for you to express your emotions. Medical device maker MindMaze has come up with a novel, compelling way to convey your facial expressions in VR called Mask. It's a foam insert that's compatible with existing headsets and uses diodes to read your biosignals and muscles. The potential applications here are plenty: You could deduce, from your opponents' faces, when they're preparing to shoot or see a new acquaintance laugh at your joke in social VR scenarios.

Researchers at MIT and Northeastern University have equipped a robot with a novel tactile sensor that lets it grasp a USB cable draped freely over a hook and insert it into a USB port. The sensor is an adaptation of a technology called GelSight, which was developed by the lab of Edward Adelson, the John and Dorothy Wilson Professor of Vision Science at MIT, and first described in 2009. The new sensor isn't as sensitive as the original GelSight sensor, which could resolve details on the micrometer scale. But it's smaller -- small enough to fit on a robot's gripper -- and its processing algorithm is faster, so it can give the robot feedback in real time. Industrial robots are capable of remarkable precision when the objects they're manipulating are perfectly positioned in advance. But according to Robert Platt, an assistant professor of computer science at Northeastern and the research team's robotics expert, for a robot taking its bearings as it goes, this type of fine-grained manipulation is unprecedented.