Abstract: Recent progress in artificial intelligence (AI) has boosted the computational possibilities in fields in which standard computers are not able to perform adequately. The AI paradigm is to emulate human intelligence and therefore breaks the familiar architecture on which digital computers are based. In particular, neuromorphic computing, artificial neural networks (ANNs), and deep learning models mimic how the brain computes. There are many applications for large networks of interconnected neurons whose synapses are individually strengthened or weakened during the learning phase. In this respect, photonics is a suitable platform for implementing ANN hardware owing to its speed, low power dissipation, and multi-wavelength opportunities. One photonic device that could serve as an optical neuron is the optical microring resonator. Indeed, microring resonators exhibit a nonlinear response and the capability for optical energy storage, which can be used to implement fading memory. In addition, their characteristic resonant behavior makes them extremely sensitive to input wavelengths, which promotes wavelength division multiplexing (WDM) applications and enables their use as WDM-based synapses (weight banks) in the linear regime. Remarkably, using silicon photonics, photonic integrated circuits can be fabricated in volume and with integrated electronics onboard. For these reasons, here, we describe the physics of silicon microring resonators and arrays of microring resonators for application in neuromorphic computing. We describe different types of ANNs, from feedforward networks to photonic extreme learning machines, and reservoir computing. In addition, we discuss hybrid systems in which silicon microresonators are coupled with other active materials. This review introduces the basics and discusses the most recent developments in the field.

Abstract: Classical random walk formalism shows a significant role across a wide range of applications. As its quantum counterpart, the quantum walk is proposed as an important theoretical model for quantum computing. By exploiting the quantum effects such as superposition, interference and entanglement, quantum walks and their variety have been extensively studied for achieving beyond classical computing power, and they have been broadly used in designing quantum algorithms in fields ranging from algebraic and optimization problems, graph and network analysis, to quantum Hamiltonian and biochemical process simulations, and even further quantum walk models have proven their capabilities for universal quantum computation. Compared to the conventional quantum circuit models, quantum walks show a feasible path for implementing application-specific quantum computing in particularly the noisy intermediate-scale quantum era. Recently remarkable progress has been achieved in implementing a wide variety of quantum walks and quantum walk applications, demonstrating the great potential of quantum walks. In this review, we provide a thorough summary of quantum walks and quantum walk computing, including aspects of quantum walk theories and characteristics, advances in their physical implementations and the flourishingly developed quantum walk computing applications. We also discuss the challenges facing quantum walk computing, toward realizing a practical quantum computer in the near future.

Abstract: Quantum machine learning (QML) is emerging as an application of quantum computing with the potential to deliver quantum advantage, but its realisation for practical applications remains impeded by challenges. Amongst those, a key barrier is the computationally expensive task of encoding classical data into a quantum state, which could erase any prospective speed-ups over classical algorithms. In this work, we implement methods for the efficient preparation of quantum states representing encoded image data using variational, genetic and matrix product state based algorithms. Our results show that these methods can approximately prepare states to a level suitable for QML using circuits two orders of magnitude shallower than a standard state preparation implementation, obtaining drastic savings in circuit depth and gate count without unduly sacrificing classification accuracy. Additionally, the QML models trained and evaluated on approximately encoded data display an increased robustness to adversarially generated input data perturbations. This partial alleviation of adversarial vulnerability, possible due to the"drowning out"of adversarial perturbations while retaining the meaningful large-scale features of the data, constitutes a considerable benefit for approximate state preparation in addition to lessening the requirements of the quantum hardware. Our results, based on simulations and experiments on IBM quantum devices, highlight a promising pathway for the future implementation of accurate and robust QML models on complex datasets relevant for practical applications, bringing the possibility of NISQ-era QML advantage closer to reality.

Abstract: Optical neural networks (ONNs) have recently attracted extensive interest as potential alternatives to electronic artificial neural networks, owing to their intrinsic capabilities in parallel signal processing with reduced power consumption and low latency. Preliminary confirmation of parallelism in optical computing has been widely performed by applying wavelength division multiplexing (WDM) to the linear transformation of neural networks. However, interchannel crosstalk has obstructed WDM technologies from being deployed in nonlinear activation on ONNs. Here, we propose a universal WDM structure called multiplexed neuron sets (MNS), which applies WDM technologies to optical neurons and enables ONNs to be further compressed. A corresponding backpropagation (BP) training algorithm was proposed to alleviate or even annul the influence of interchannel crosstalk in MNS-based WDM-ONNs. For simplicity, semiconductor optical amplifiers are employed as an example of MNS to construct a WDM-ONN trained using the new algorithm. The results show that the combination of MNS and the corresponding BP training algorithm clearly downsizes the system and improves the energy efficiency by a factor of 10 while providing similar performance to traditional ONNs.

Abstract: A Kerr nonlinear oscillator (KNO) supports a pair of steady eigenstates, coherent states with opposite phases, that are good for the encoding of continuous variable qubit basis states.Arbitrary control of the KNO confined within the steady state subspace allows extraction of the Berry curvature through the linear response of the physical observable to the quench velocity of the system, providing an effective method for the characterization of topology in the KNO.As an alternative, the control adopting the "shortcut to adiabaticity" to the KNO enables the exploration of the topology through accelerated adiabatic eigenstate evolution to measure all three physical observables.Topological transitions are revealed by the jump of the first Chern number, obtained respectively from the integral of the Berry curvature and of the new polar angle relation, over the whole parameter space.Our strategy paves the way for measuring topological transitions in continuous variable systems.

Abstract: The advent of neuromorphic electronics is increasingly revolutionizing the concept of computation. In the last decade, several studies have shown how materials, architectures, and neuromorphic devices can be leveraged to achieve brain-like computation with limited power consumption and high energy efficiency. Neuromorphic systems have been mainly conceived to support spiking neural networks that embed bioinspired plasticity rules such as spike time-dependent plasticity to potentially support both unsupervised and supervised learning. Despite substantial progress in the field, the information transfer capabilities of biological circuits have not yet been achieved. More importantly, demonstrations of the actual performance of neuromorphic systems in this context have never been presented. In this paper, we report similarities between biological, simulated, and artificially reconstructed microcircuits in terms of information transfer from a computational perspective. Specifically, we extensively analyzed the mutual information transfer at the synapse between mossy fibers and granule cells by measuring the relationship between pre- and post-synaptic variability. We extended this analysis to memristor synapses that embed rate-based learning rules, thus providing quantitative validation for neuromorphic hardware and demonstrating the reliability of brain-inspired applications.

Abstract: Speech emotion recognition (SER) is constantly gaining attention in recent years due to its potential applications in diverse fields and thanks to the possibility offered by deep learning technologies. However, recent studies have shown that deep learning models can be vulnerable to adversarial attacks. In this paper, we systematically assess this problem by examining the impact of various adversarial white-box and black-box attacks on different languages and genders within the context of SER. We first propose a suitable methodology for audio data processing, feature extraction, and CNN-LSTM architecture. The observed outcomes highlighted the significant vulnerability of CNN-LSTM models to adversarial examples (AEs). In fact, all the considered adversarial attacks are able to significantly reduce the performance of the constructed models. Furthermore, when assessing the efficacy of the attacks, minor differences were noted between the languages analyzed as well as between male and female speech. In summary, this work contributes to the understanding of the robustness of CNN-LSTM models, particularly in SER scenarios, and the impact of AEs. Interestingly, our findings serve as a baseline for a) developing more robust algorithms for SER, b) designing more effective attacks, c) investigating possible defenses, d) improved understanding of the vocal differences between different languages and genders, and e) overall, enhancing our comprehension of the SER task.

Abstract: The world has seen the emergence of machines based on pretrained models, transformers, also known as generative artificial intelligences for their ability to produce various types of content, including text, images, audio, and synthetic data. Without resorting to preprogramming or special tricks, their intelligence grows as they learn from experience, and to ordinary people, they can appear human-like in conversation. This means that they can pass the Turing test, and that we are now living in one of many possible Turing futures where machines can pass for what they are not. However, the learning machines that Turing imagined would pass his imitation tests were machines inspired by the natural development of the low-energy human cortex. They would be raised like human children and naturally learn the ability to deceive an observer. These ``child machines,'' Turing hoped, would be powerful enough to have an impact on society and nature.