Next-generation Artificial Intelligence Research Articles

Firing feature-driven neural circuits with scalable memristive neurons for robotic obstacle avoidance

Neural circuits with specific structures and diverse neuronal firing features are the foundation for supporting intelligent tasks in biology and are regarded as the driver for catalyzing next-generation artificial intelligence. Emulating neural circuits in hardware underpins engineering highly efficient neuromorphic chips, however, implementing a firing features-driven functional neural circuit is still an open question. In this work, inspired by avoidance neural circuits of crickets, we construct a spiking feature-driven sensorimotor control neural circuit consisting of three memristive Hodgkin-Huxley neurons. The ascending neurons exhibit mixed tonic spiking and bursting features, which are used for encoding sensing input. Additionally, we innovatively introduce a selective communication scheme in biology to decode mixed firing features using two descending neurons. We proceed to integrate such a neural circuit with a robot for avoidance control and achieve lower latency than conventional platforms. These results provide a foundation for implementing real brain-like systems driven by firing features with memristive neurons and put constructing high-order intelligent machines on the agenda.

Evaluation and Improvement Strategies of Teachers’ Information Technology Teaching Ability Based on Data Mining

The rapid evolution of emerging technologies such as the Internet of Things, big data analytics, and next-generation artificial intelligence has propelled education through the ranks to new heights. To promote the development of digital education in an all-round way and build a learning country under the strategic action of digital education in China, it is necessary to objectively and accurately evaluate the information-based teaching ability of college teachers. This paper takes the evaluation of teachers’ information-based teaching ability and the construction of promotion strategies as the research object, introduces the composition of teachers’ information-based teaching ability in detail, designs the evaluation index system of teachers’ information-based teaching ability from the perspective of educational digital strategic action, and constructs the promotion strategies of teachers’ information-based teaching ability based on the case study of teaching ability evaluation. The constructed evaluation index of teachers’ informatization teaching ability has strong maneuverability and the evaluation results are objective and accurate, which can be used as an evaluation model in the construction of teachers’ informatization teaching ability in colleges and universities and provide practical guidance for improving teachers’ informatization teaching ability in colleges and universities.

Knowledge engineering for wind energy

Abstract. With the rapid evolution of the wind energy sector, there is an ever-increasing need to create value from the vast amounts of data made available both from within the domain and from other sectors. This article addresses the challenges faced by wind energy domain experts in converting data into domain knowledge, connecting and integrating them with other sources of knowledge, and making them available for use in next-generation artificial intelligence systems. To this end, this article highlights the role that knowledge engineering can play in the digital transformation of the wind energy sector. It presents the main concepts underpinning knowledge-based systems and summarises previous work in the areas of knowledge engineering and knowledge representation in a manner that is relevant and accessible to wind energy domain experts. A systematic analysis of the current state of the art on knowledge engineering in the wind energy domain is performed with available tools put into perspective by establishing the main domain actors and their needs, as well as identifying key problematic areas. Finally, recommendations for further development and improvement are provided.

Coupling thermogalvanic and piezoresistive effects in a robust hydrogel for Deep-Learning-Assisted Self-Powered sign language and object recognition

The human hand provides rich sensory information for proper interaction with the environment. However, designing passive hand-sensing arrays with robust mechanical properties, high conductivity, ease of integration and reliability remains a long-term challenge. Here, we develop a robust dynamic crystalline ion-injected thermoelectric gel (DCI-TEG) via a simple solvent-exchange salt-out strategy. The resultant DCI-TEG exhibits high tensile strength (3.8 MPa), toughness (7.5 MJ m−3), and ionic conductivity of 1.84 S m−1. Benefiting from non-covalent crosslinking, the gel electrolyte and electrodes can be fully integrated into modules with tight physical and electrical contact, facilitating a stable and reliable signal output. By coupling thermogalvanic and piezoresistive effects, the DCI-TEG permits self-powered timely sensing of pressure using current signals. By implanting ionic thermoelectric module arrays on a glove, a wearable pressure sensing route is demonstrated with the merits of self-power, low cost and simple preparation. With the help of deep learning algorithms, a self-powered recognition for sign language and objects is realized in wearable/portable scenarios by active high-accuracy pressure detection. This work gives people a new direction toward barrier-free passive communication and human–machine interaction in the next-generation artificial intelligence.

A Liver Cancer Question-Answering System Based on Next-Generation Intelligence and the Large Model Med-PaLM 2

Chronic diseases are the "number one killer" threatening human life and health. In recent years, the incidence of chronic diseases has been rising, and the trend is younger, and the prevention and control situation is very serious. This study introduces a Liver Cancer Question-Answering System (LCQAS) that leverages next-generation artificial intelligence and the large model Med-PaLM 2 The Med-PaLM 2 medical question answering system, powered by next-generation intelligence and large language models, represents a cutting-edge tool in the healthcare domain. Leveraging advanced artificial intelligence techniques and the capabilities of large language models like Med-PaLM 2, this system aims to provide accurate and comprehensive responses to medical inquiries and queries. With its intuitive interface and contextually relevant answers, LCQAS represents a significant advancement in medical information retrieval for liver cancer management.

Edge Offloading in Smart Grid

The management of decentralized energy resources and smart grids needs novel data-driven low-latency applications and services to improve resilience and responsiveness and ensure closer to real-time control. However, the large-scale integration of Internet of Things (IoT) devices has led to the generation of significant amounts of data at the edge of the grid, posing challenges for the traditional cloud-based smart-grid architectures to meet the stringent latency and response time requirements of emerging applications. In this paper, we delve into the energy grid and computational distribution architectures, including edge–fog–cloud models, computational orchestration, and smart-grid frameworks to support the design and offloading of grid applications across the computational continuum. Key factors influencing the offloading process, such as network performance, data and Artificial Intelligence (AI) processes, computational requirements, application-specific factors, and energy efficiency, are analyzed considering the smart-grid operational requirements. We conduct a comprehensive overview of the current research landscape to support decision-making regarding offloading strategies from cloud to fog or edge. The focus is on metaheuristics for identifying near-optimal solutions and reinforcement learning for adaptively optimizing the process. A macro perspective on determining when and what to offload in the smart grid is provided for the next-generation AI applications, offering an overview of the features and trade-offs for selecting between federated learning and edge AI solutions. Finally, the work contributes to a comprehensive understanding of edge offloading in smart grids, providing a Strengths, Weaknesses, Opportunities, and Threats (SWOT) analysis to support cost–benefit analysis in decision-making regarding offloading strategies.

Monolithically Integrated Complementary Ferroelectric FET XNOR Synapse for the Binary Neural Network.

Neuromorphic computing, which mimics the structure and principles of the human brain, has the potential to facilitate the hardware implementation of next-generation artificial intelligence systems and process large amounts of data with very low power consumption. Among them, the XNOR synapse-based Binary Neural Network (BNN) has been attracting attention due to its compact neural network parameter size and low hardware cost. The previous XNOR synapse has drawbacks, such as a trade-off between cell density and accuracy. In this work, we show nonvolatile XNOR synapses with high density and accuracy using a monolithically stacked complementary ferroelectric field-effect transistor (C-FeFET) composed of a p-type Si MFMIS-FeFET at the bottom and a 3D stackable n-type Al:IZTO MFS-FeTFT, achieving 60F2 per cell (2C-FeFET). For adjusting the threshold voltage and improving the switching speed (100 ns) of n-type ferroelectric TFT, we employed a dual-gate configuration and a unique operation scheme, making it comparable to those of Si-based FeFETs. We performed array-level simulation with a 512 × 512 subarray size and a 3-bit flash ADC, demonstrating that the image recognition accuracies using the MNIST and CIFAR-10 data sets were increased by 3.17 and 14.07%, respectively, in comparison to other nonvolatile XNOR synapses. In addition, we performed system-level analysis on a 512 × 512 XNOR C-FeFET, exhibiting an outstanding throughput of 717.37 GOPS and an energy efficiency of 196.7 TOPS/W. We expect that our approach would contribute to the high-density memory systems, logic-in-memory technology, and hardware implementation of neural networks.

Memristor-Based Artificial Chips.

Memristors, promising nanoelectronic devices with in-memory resistive switching behavior that is assembled with a physically integrated core processing unit (CPU) and memory unit and even possesses highly possible multistate electrical behavior, could avoid the von Neumann bottleneck of traditional computing devices and show a highly efficient ability of parallel computation and high information storage. These advantages position them as potential candidates for future data-centric computing requirements and add remarkable vigor to the research of next-generation artificial intelligence (AI) systems, particularly those that involve brain-like intelligence applications. This work provides an overview of the evolution of memristor-based devices, from their initial use in creating artificial synapses and neural networks to their application in developing advanced AI systems and brain-like chips. It offers a broad perspective of the key device primitives enabling their special applications from the view of materials, nanostructure, and mechanism models. We highlight these demonstrations of memristor-based nanoelectronic devices that have potential for use in the field of brain-like AI, point out the existing challenges of memristor-based nanodevices toward brain-like chips, and propose the guiding principle and promising outlook for future device promotion and system optimization in the biomedical AI field.

Memristor-Based Bionic Tactile Devices: Opening the Door for Next-Generation Artificial Intelligence.

Bioinspired tactile devices can effectively mimic and reproduce the functions of the human tactile system, presenting significant potential in the field of next-generation wearable electronics. In particular, memristor-based bionic tactile devices have attracted considerable attention due to their exceptional characteristics of high flexibility, low power consumption, and adaptability. These devices provide advanced wearability and high-precision tactile sensing capabilities, thus emerging as an important research area within bioinspired electronics. This paper delves into the integration of memristors with other sensing and controlling systems and offers a comprehensive analysis of the recent research advancements in memristor-based bionic tactile devices. These advancements incorporate artificial nociceptors and flexible electronic skin (e-skin) into the category of bio-inspired sensors equipped with capabilities for sensing, processing, and responding to stimuli, which are expected to catalyze revolutionary changes in human-computer interaction. Finally, this review discusses the challenges faced by memristor-based bionic tactile devices in terms of material selection, structural design, and sensor signal processing for the development of artificial intelligence. Additionally, it also outlines future research directions and application prospects of these devices, while proposing feasible solutions to address the identified challenges.

NDR effect and negative-set/reset phenomenon in core-shell structure-based memristor under the effect of humidity

The potential applications of memristors in the post-Moore era of rapidly developing next-generation artificial intelligence and high-frequency communication technologies are immense. The effects on the resistive switching (RS) characteristics based on Ag Salt@TiO2 memristive device with a nano core-shell structure under different relative humidity (RH) environments have been investigated in this work. Negative-set/reset phenomena occurs in negative voltage region due to excessive growth of conductive filaments (CFs). When the CF returns to normal growth, the current suddenly decreases in the positive high voltage area at lower humidity, which can be explained by the dissolution and growth hindrance of the CFs. Furthermore, the coexistence of negative differential resistance (NDR) and bipolar RS at room temperature can be regulated by adjusting humidity. The current conduction mechanism of the device in both dry and humid environments were fitted using the ohmic conduction mechanism, space charge-limited current (SCLC) conduction mechanism, and Schottky emission mechanism.

Nanowire-Enhanced Fully Transparent and Flexible Indium Gallium Zinc Oxide Transistors with Chitosan Hydrogel Gate Dielectric: A Pathway to Improved Synaptic Properties.

In this study, a transparent and flexible synaptic transistor was fabricated based on a random-network nanowire (NW) channel made of indium gallium zinc oxide. This device employs a biocompatible chitosan-based hydrogel as an electrolytic gate dielectric. The NW structure, with its high surface-to-volume ratio, facilitated a more effective modulation of the channel conductance induced by protonic-ion polarization. A comparative analysis of the synaptic properties of NW- and film-type devices revealed the distinctive features of the NW-type configuration. In particular, the NW-type synaptic transistors exhibited a significantly larger hysteresis window under identical gate-bias conditions. Notably, these transistors demonstrated enhanced paired-pulse facilitation properties, synaptic weight modulation, and transition from short- to long-term memory. The NW-type devices displayed gradual potentiation and depression of the channel conductance and thus achieved a broader dynamic range, improved linearity, and reduced power consumption compared with their film-type counterparts. Remarkably, the NW-type synaptic transistors exhibited impressive recognition accuracy outcomes in Modified National Institute of Standards and Technology pattern-recognition simulations. This characteristic enhances the efficiency of practical artificial intelligence (AI) processes. Consequently, the proposed NW-type synaptic transistor is expected to emerge as a superior candidate for use in high-efficiency artificial neural network systems, thus making it a promising technology for next-generation AI semiconductor applications.

Enhanced Synaptic Behaviors in Chitosan Electrolyte-Based Electric-Double-Layer Transistors with Poly-Si Nanowire Channel Structures.

In this study, we enhance the synaptic behavior of artificial synaptic transistors by utilizing nanowire (NW)-type polysilicon channel structures. The high surface-to-volume ratio of the NW channels enables efficient modulation of the channel conductance, which is interpreted as the synaptic weight. As a result, NW-type synaptic transistors exhibit a larger hysteresis window compared to film-type synaptic transistors, even within the same gate voltage sweeping range. Moreover, NW-type synaptic transistors demonstrate superior short-term facilitation and long-term memory transition compared with film-type ones, as evidenced by the measured paired-pulse facilitation and excitatory post-synaptic current characteristics at varying frequencies and pulse numbers. Additionally, we observed gradual potentiation/depression characteristics, making these artificial synapses applicable to artificial neural networks. Furthermore, the NW-type synaptic transistors exhibit improved Modified National Institute of Standards and Technology pattern recognition rate of 91.2%. In conclusion, NW structure channels are expected to be a promising technology for next-generation artificial intelligence (AI) semiconductors, and the integration of NW structure channels has significant potential to advance AI semiconductor technology.

The Role of ChatGPT in Improving Cross-Cultural Team Management Performance

In this era of advanced globalisation, intercultural collaboration has become commonplace in business and organisational environments. Companies and organisations often form cross-cultural teams consisting of individuals from different cultural backgrounds, languages and values. These cross-cultural teams have great potential to provide a competitive advantage due to the combination of diverse knowledge and perspectives, but can also face challenges in effective team management and performance. This research aims to explore and explore the potential of ChatGPT utilisation in improving cross-cultural team management performance. The current research type is qualitative. Data collection techniques include listening and recording important information to conduct data analysis through data reduction, data display, and conclusion drawing. The study results show that the use of ChatGPT in cross-cultural team management brings great benefits to companies and organisations operating in a global business environment. As a form of next-generation Artificial Intelligence, ChatGPT can facilitate cross-cultural communication, increase understanding of cultural differences, and help better manage conflict. Features such as language translation, sentiment analysis, and intercultural skills training contribute to more efficient and creative team performance.

Pure-Attention-Based Multifunction Memristive Neuromorphic Circuit and System

The use of memristive neuromorphic circuit and system is a promising solution for next-generation Artificial Intelligence (AI) computing, as it offers possibilities that go beyond conventional GPU-based artificial neural network computing platforms. However, most of the existing memristive neuromorphic circuits and systems are designed for the specific networks, which is lack of universality and flexibility. Therefore, this paper proposes a universal memristive circuit and system framework for pure-attention-based transformer networks to implement multifunction applications on edge devices. Furthermore, the verification of image recognition and speech recognition was achieved by extending the size of the memristor crossbar array macros and reconfiguring the memristor weights without changing the memristive transformer circuit and framework. This paper not only provides a universal edge implementation framework for multifunction applications of the transformer, but also offers a low-power and promising solution for the application of pure-attention-based transformers on edge devices.

Socially intelligent machines that learn from humans and help humans learn.

A hallmark of human intelligence is the ability to understand and influence other minds. Humans engage in inferential social learning (ISL)by using commonsense psychology to learn from others and help others learn. Recent advances in artificial intelligence (AI) are raising new questions about the feasibility of human-machine interactions that support such powerful modes of social learning. Here, we envision what it means to develop socially intelligent machines that can learn, teach, and communicate in ways that are characteristic of ISL. Rather than machines that simply predict human behaviours or recapitulate superficial aspects of human sociality (e.g. smiling, imitating), we should aim to build machines that can learn from human inputs and generate outputs for humans byproactively considering human values, intentions and beliefs. While such machines can inspire next-generation AI systems that learn more effectively from humans (as learners) and even help humans acquire new knowledge (as teachers), achieving these goals will also require scientific studies of its counterpart: how humans reason about machine minds and behaviours. We close by discussing the need for closer collaborations between the AI/ML and cognitive science communities to advance a science of both natural and artificial intelligence. This article is part of a discussion meeting issue 'Cognitive artificial intelligence'.

Super-Flexible, Transparent Synaptic Transistors Based on Pullulan for Neuromorphic Electronics

With the development of artificial intelligence, super-flexibility and transparent neuromorphic electronics are gradually showing high application value in the fields of wearable see-through electronics and biomedical applications. However, it is still a huge problem to realize synaptic devices with super-flexibility and transparent characteristics. Here, an organic synaptic transistor (OST) based on natural pullulan is fabricated with super-flexibility and high transmittance (~83.3%), which enables it to achieve conformal and non-destructive attachment to arbitrary-shaped objects without affecting viewing. Various synaptic behaviors are successfully simulated even under extreme bending, showing outstanding mechanical stability. Moreover, our OSTs can achieve ultrarapid degradation in ambient water, which will reduce the generation of electronic waste. This work provides a guide for the development of super-flexible and transparent neuromorphic electrons, demonstrating a great step toward next-generation artificial intelligence.

Short-term synaptic plasticity in emerging devices for neuromorphic computing.

Neuromorphic computing is a promising computing paradigm toward building next-generation artificial intelligence machines, in which diverse types of synaptic plasticity play an active role in information processing. Compared to long-term plasticity (LTP) forming the foundation of learning and memory, short-term plasticity (STP) is essential for critical computational functions. So far, the practical applications of LTP have been widely investigated, whereas the implementation of STP in hardware is still elusive. Here, we review the development of STP by bridging the physics in emerging devices and biological behaviors. We explore the computational functions of various STP in biology and review their recent progress. Finally, we discuss the main challenges of introducing STP into synaptic devices and offer the potential approaches to utilize STP to enrich systems' capabilities. This review is expected to provide prospective ideas for implementing STP in emerging devices and may promote the construction of high-level neuromorphic machines.

ZTO/MgO-Based Optoelectronic Synaptic Memristor for Neuromorphic Computing

Synapse having good linearity plays a vital role in the memory and computing of human brain. Therefore, the achievement of efficient learning process in neuromorphic computing by the implementation of the synaptic functions, such as long-term potentiation/ depression (LTP/LTD) and spike time-dependent plasticity (STDP) of two-terminal optoelectronic memristor device, is critical for the next-generation artificial intelligence. In this work, we improve the resistive switching and synaptic characteristics of a Zn2SnO4 (ZTO)-based optoelectronic synaptic memristor (OSM) by the insertion of an ultrathin MgO layer. For this bilayer (BL) structured OSM, the nonlinearities of LTP and LTD curves are improved to 1.96 and 0.33, respectively. Asymmetrical STDP response demonstrates the suitability of device toward the Hebbian learning. In addition, a Hopfield neural network (HNN) is successfully trained to recognize a 10 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\times10$ </tex-math></inline-formula> pixel input image with an accuracy of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim 100$ </tex-math></inline-formula> % after 15 iterations. Under blue light (405 nm) illumination, OSM emulates the synaptic functions, such as paired pulse facilitation, learning experience behavior, and short- to long-term memory transition. The photoresponse and relaxation characteristics of the device depend on the ionization and neutralization of oxygen vacancies. This highly transparent ZTO/MgO-based OSM with the convergence of “nonvolatile electronic memory and visible light sensor” is suitable as an artificial synapse for neuromorphic computing applications.

SqueezeLight: A Multi-Operand Ring-Based Optical Neural Network With Cross-Layer Scalability

Optical neural networks (ONNs) are promising hardware platforms for next-generation artificial intelligence acceleration with ultrafast speed and low-energy consumption. However, previous ONN designs are bounded by one multiply–accumulate operation per device, showing unsatisfying scalability. In this work, we propose a scalable ONN architecture, dubbed <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SqueezeLight</monospace> . We propose a nonlinear optical neuron based on multioperand ring resonators (MORRs) to squeeze vector dot-product into a single device with low wavelength usage and built-in nonlinearity. A block-level squeezing technique with structured sparsity is exploited to support higher scalability. We adopt a robustness-aware training algorithm to guarantee variation tolerance. To enable a truly scalable ONN architecture, we extend <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SqueezeLight</monospace> to a separable optical CNN architecture that further squeezes in the layer level. Two orthogonal convolutional layers are mapped to one MORR array, leading to order-of-magnitude higher software training scalability. We further explore augmented representability for <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SqueezeLight</monospace> by introducing parametric MORR neurons with trainable nonlinearity, together with a nonlinearity-aware initialization method to stabilize convergence. Experimental results show that <monospace xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SqueezeLight</monospace> achieves one-order-of-magnitude better compactness and efficiency than previous designs with high fidelity, trainability, and robustness. Our open-source codes are available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/JeremieMelo/SqueezeLight</uri> .

Artificial intelligence chatbots will revolutionize how cancer patients access information: ChatGPT represents a paradigm-shift.

On November 30, 2022, OpenAI enabled public access to ChatGPT, a next-generation artificial intelligence with a highly sophisticated ability to write, solve coding issues, and answer questions. This communication draws attention to the prospect that ChatGPT and its successors will become important virtual assistants to patients and health-care providers. In our assessments, ranging from answering basic fact-based questions to responding to complex clinical questions, ChatGPT demonstrated a remarkable ability to formulate interpretable responses, which appeared to minimize the likelihood of alarm compared with Google's feature snippet. Arguably, the ChatGPT use case presents an urgent need for regulators and health-care professionals to be involved in developing standards for minimum quality and to raise patient awareness of current limitations of emerging artificial intelligence assistants. This commentary aims to raise awareness at the tipping point of a paradigm shift.