Flexible Network Research Articles

Aberrant Flexibility of Dynamic Brain Network in Patients with Autism Spectrum Disorder.

Autism spectrum disorder (ASD) is a collection of neurodevelopmental disorders whose pathobiology remains elusive. This study aimed to investigate the possible neural mechanisms underlying ASD using a dynamic brain network model and a relatively large-sample, multi-site dataset. Resting-state functional magnetic resonance imaging data were acquired from 208 ASD patients and 227 typical development (TD) controls, who were drawn from the multi-site Autism Brain Imaging Data Exchange (ABIDE) database. Brain network flexibilities were estimated and compared between the ASD and TD groups at both global and local levels, after adjusting for sex, age, head motion, and site effects. The results revealed significantly increased brain network flexibilities (indicating a decreased stability) at the global level, as well as at the local level within the default mode and sensorimotor areas in ASD patients than TD participants. Additionally, significant ASD-related decreases in flexibilities were also observed in several occipital regions at the nodal level. Most of these changes were significantly correlated with the Autism Diagnostic Observation Schedule (ADOS) total score in the entire sample. These results suggested that ASD is characterized by significant changes in temporal stabilities of the functional brain network, which can further strengthen our understanding of the pathobiology of ASD.

Fault Localization in Multi-Terminal DC Distribution Networks Based on PSO Algorithm

Flexible DC power grids are widely recognized as an important component of building smart grids. Compared with traditional AC power grids, flexible DC power grids have strong technical advantages in islanding power supplies, distributed power supplies, regional power supplies, and AC system interconnection. In multi-terminal flexible DC power grids containing renewable energy sources such as solar and wind power, due to the instability and intermittency of renewable energy, it is usually necessary to add energy storage units to pre-regulate the power of the multi-terminal flexible DC power grid in islanded operation. Aiming at the important problem of large current impact and serious consequences when the flexible DC distribution network fails, a combined location method combining an improved impedance method (series current-limiting reactors at both ends of the line to obtain a more accurate current differential value) and a particle swarm optimization algorithm is proposed. Initially, by establishing the enhanced impedance model, the differential variables under the conditions of inter-electrode short-circuit and single-pole grounding fault can be obtained. Then tailor-made fitness functions are designed for these two models to optimize parameter identification. Subsequently, the iterative parameters of the particle swarm optimization algorithm are fine-tuned, giving it dynamic sociality and self-learning ability in the iterative process, which significantly improves the convergence speed and successfully avoids local optimization. Finally, various fault types in a six-terminal DC distribution network are simulated and analyzed by MATLAB, and the results show that this method has good accuracy and robustness. This research provides strong theoretical and methodological support for improving the safety and reliability of DC distribution systems.

A Two-Stage Hybrid Stochastic–Robust Coordination of Combined Energy Management and Self-Healing in Smart Distribution Networks Incorporating Multiple Microgrids

This paper presents a two-stage hybrid stochastic–robust coordination of energy management and self-healing in smart distribution networks with multiple microgrids. A multi-agent systems approach is first used for coupling energy management and self-healing strategies of microgrids, based on expert system rules. The second stage problem, a framework similar to that of the first stage, is then established for the smart distribution networks. Then, hybrid stochastic–robust optimization is used to model the uncertainties of demand, energy price, power generation of renewable energy sources, demand of electric vehicles, and accessibility of zone agents. Further, the grey wolf algorithm is used to solve the formulated optimization problem and achieve an optimal and reliable solution. The proposal is validated on a 69-bus distribution network consisting of three microgrids. The results validate that the proposal minimizes microgrids’ utilization indices, such as energy costs, energy losses, and network voltage drops, while simultaneously managing a flexible distribution network. It is also verified that the proposed multi-agent system design provides a high-speed and optimized self-healing solution for the network.

A novel approach to synthesize micrometer-sized porous Si/SiOx@C with internally and externally interweaved CNTs as a high performance anode for lithium-ion batteries

Silicon-based materials are prospective anodes because of the high capability of lithium (Li) storage, but the large volumetric expansion seriously destroys the integrity of the electrodes and further results in unsatisfactory cycling performance. Reasonable structure design based on low-cost nano silicon (Si) prepared by an feasible method is still an enormous challenge. Herein, the carbon-coated and cross-linked carbon nanotube/porous-Si/SiOx microspheres (CNT/SFDP-Si/SiOx@C) are firstly synthesized by combining an inverse water-in-oil microemulsion approach with magnesiothermic reduction. One high value-added route for utilizing silica fume as the precursor to form nano silicon is provided. The collaborative strategy via wired CNTs network, porous-Si/SiOx flexible matrix, and carbon coating is employed to keep the integrity and stability of electrode structure. Benefiting from the synergistic effect of the flexibility of CNTs network, protection of outer carbon shell, doping of diverse nitrogen (N) species, and accommodation of porous-Si/SiOx, CNT/SFDP-Si/SiOx@C exhibits excellent electrochemical performance. The long cycling lifespan with a reversible capacity of 675.8 mAh g−1 after 1000 cycles at 2 A g−1, a slight decay rate per cycle of 0.018 %, and a high-rate performance are simultaneously obtained. The prelithiated LiFePO4||CNT/SFDP-Si/SiOx@C full cell is also assembled, showing the reversible capacity of 129.4 mAh g−1 at 0.2C.

A Survey an Optimized Dense CNN Model for Recognizing Deepfake Images

Deepfake detection relies on a deep learning model. Deepfake content is generated using artificial intelligence and machine learning techniques to swap one person’s face with another’s in images. These altered images are certain to have a significant impact on society. Deepfakes utilize advanced technologies such as machine learning (ML) and deep learning (DL) to develop automated techniques for generating deceptive content. The model is trained using the DFDC (Deepfake Detection Challenge) dataset, which includes 1, 00,000 videos comprising both real and fake content. This survey examines recent progress in deep learning techniques for detecting deepfake images, highlighting their growing significance in today’s digital landscape. The paper outlines the issues arising from deepfakes, explores the techniques for generating deceptive content, and analyzes various deep learning models employed in deepfake detection, such as convolutional neural networks (CNN’s), MTCNN (Multi-task CNN), and Facial Landmark. It emphasizes the ongoing battle between the progression of deepfakes and the evolution of detection methods, underscoring the need for advanced and flexible neural network architectures to effectively curb the dissemination of deceptive information.

A robust adapted Flexible Parallel Neural Network architecture for early prediction of lithium battery lifespan

Early prediction of End of Life (EPEOL) is crucial for improving lithium battery efficiency and lifespan. Traditional fixed-architecture neural networks often suffer from underfitting or overfitting due to diverse data distributions. To address this, we propose the Flexible Parallel Neural Network (FPNN). This model integrates modules like InceptionBlock, 3D CNN, 2D CNN, and dual-stream networks. By effectively extracting electrochemical features from video-format data through the 3D CNN and achieving multi-scale feature abstraction with InceptionBlock, FPNN ensures effective module coordination. The model can adaptively adjust the number of InceptionBlocks to handle tasks of varying complexity. Experimental results on the MIT dataset show that FPNN achieves MAPE values of 1.26%, 0.41%, 0.37%, 0.33%, 0.32%, 0.32%, 0.31%, 0.31%, 0.22%, and 0.34% using the first 10, 20, 30, 40, 50, 60, 70, 80, 90, and 100 cycle data, respectively. The interpretability of FPNN is reflected in its structural design and flexible unit choices, providing a basis for model interpretation. Comprising multiple modules, FPNN enables smooth feature extraction from electrochemical data through collaborative interaction. The diverse branching structure of the flexible units allows the model to capture features at different scales, learning richer information. Our approach offers a precise, adaptable, and easy-to-understand solution for EPEOL, opening new possibilities in battery health monitoring.

Ubiquitin-driven protein condensation stabilizes clathrin-mediated endocytosis.

Clathrin-mediated endocytosis is an essential cellular pathway that enables signaling and recycling of transmembrane proteins and lipids. During endocytosis, dozens of cytosolic proteins come together at the plasma membrane, assembling into a highly interconnected network that drives endocytic vesicle biogenesis. Recently, multiple groups have reported that early endocytic proteins form flexible condensates, which provide a platform for efficient assembly of endocytic vesicles. Given the importance of this network in the dynamics of endocytosis, how might cells regulate its stability? Many receptors and endocytic proteins are ubiquitylated, while early endocytic proteins such as Eps15 contain ubiquitin-interacting motifs. Therefore, we examined the influence of ubiquitin on the stability of the early endocytic protein network. In vitro, we found that recruitment of small amounts of polyubiquitin dramatically increased the stability of Eps15 condensates, suggesting that ubiquitylation could nucleate endocytic assemblies. In live-cell imaging experiments, a version of Eps15 that lacked the ubiquitin-interacting motif failed to rescue defects in endocytic initiation created by Eps15 knockout. Furthermore, fusion of Eps15 to a deubiquitylase enzyme destabilized nascent endocytic sites within minutes. In both in vitro and live-cell settings, dynamic exchange of Eps15 proteins, a measure of protein network stability, was decreased by Eps15-ubiquitin interactions and increased by loss of ubiquitin. These results collectively suggest that ubiquitylation drives assembly of the flexible protein network responsible for catalyzing endocytic events. More broadly, this work illustrates a biophysical mechanism by which ubiquitylated transmembrane proteins at the plasma membrane could regulate the efficiency of endocytic internalization.

A Vehicle-to-Grid planning framework incorporating electric vehicle user equilibrium and distribution network flexibility enhancement

The rapid surge in electric vehicle (EV) adoption, coupled with advancements in charging technologies, emphasizes the critical necessity for expanding EV recharging infrastructure. Simultaneously, the Distribution Network (DN) encounters escalating challenges in meeting charging demand during peak traffic periods. Consequently, there is a mounting demand for the deployment of innovative Vehicle-to-Grid (V2G) technologies to augment the DN’s flexibility in power dispatch and alleviate travel costs for EV users. Hence, this paper proposes an EV-user-equilibrium-(UE)-constrained V2G planning framework that enhances flexibility in the DN. The framework aims to ascertain the optimal placement and capacity of EV charging stations (EVCSs) and V2G charging piles within the Transportation Network (TN). It takes into account the equilibrium condition stemming from competitive EV charging and routing behaviors alongside the optimal expansion of DN energy resources to accommodate the electricity supplied by the V2G piles. This study commences by analyzing EV drivers’ travel decisions, considering the influence of charging and V2G pile locations and sizes. Subsequently, we tackle the Traffic Assignment Problem with User Equilibrium (TAP-UE) model to characterize the steady-state traffic flow distribution of EVs. Following this, we formulate the optimization model for the Coordinated Power and Transportation Network (CPTN), which encompasses the optimal expansion of DN facilities and traffic flow regulation under UE conditions. To mitigate the computational complexity associated with the V2G planning model, we introduce a series of linearization methods to obtain a manageable Mixed-Integer Linear Programming (MILP) solution. Finally, to validate the efficacy of our proposed planning framework, we apply it to two test systems, including a real-world case study. Through these case studies, we explore the necessity and potential benefits of V2G technologies.

Rational Construction of Honeycomb-like Carbon Network-Encapsulated MoSe2 Nanocrystals as Bifunctional Catalysts for Highly Efficient Water Splitting.

The scalable fabrication of cost-efficient bifunctional catalysts with enhanced hydrogen evolution reaction (HER) and oxygen evolution reaction (OER) performance plays a significant role in overall water splitting in hydrogen production fields. MoSe2 is considered to be one of the most promising candidates because of its low cost and high catalytic activity. Herein, hierarchical nitrogen-doped carbon networks were constructed to enhance the catalytic activity of the MoSe2-based materials by scalable free-drying combined with an in situ selenization strategy. The rationally designed carbonaceous network-encapsulated MoSe2 composite (MoSe2/NC) endows a continuous honeycomb-like structure. When utilized as a bifunctional electrocatalyst for both HER and OER, the MoSe2/NC electrode exhibits excellent electrochemical performance. Significantly, the MoSe2/NC‖MoSe2/NC cells require a mere 1.5 V to reach a current density of 10 mA cm-2 for overall water splitting in 1 M KOH. Ex situ characterizations and electrochemical kinetic analysis reveal that the superior catalytic performance of the MoSe2/NC composite is mainly attributed to fast electron and ion transportation and good structural stability, which is derived from the abundant active sites and excellent structural flexibility of the honeycomb-like carbon network. This work offers a promising pathway to the scalable fabrication of advanced non-noble bifunctional electrodes for highly efficient hydrogen evolution.

Microbial community structure and function in a sandy soil column under dynamic hydrologic regimes

The capillary fringe (CF) exhibits steep chemical and redox gradients, representing a critical reactive interface in the subsurface. The spatial extent of the CF is temporally dynamic because of water table and soil moisture fluctuations. To simulate the spatio-temporal responses of the CF’s microbial community structure and activity, a 91 cm long sandy soil column experiment was conducted by imposing three different sequential hydrologic regimes: a stable water table separating the unsaturated and saturated zones (static regime), periodic pulse infiltration events with a constant water table position (infiltration regime), and water table fluctuations under three drainage-imbibition cycles of 6, 12, and 18 days (fluctuating regime). The investigation focused on the microbially-mediated turnover of nitrogen (N) and carbon (C). Geochemical profiles in the soil column were monitored continuously, while microbial DNA was extracted for 16S rDNA sequencing at the start, middle and end of the experiment. The observed microbial community compositions corresponded to five categories: (1) initial soil, (2) near-surface and (3) CF and saturated soil under the infiltration regime, (4) unsaturated and (5) saturated soil in the fluctuating regime. Based on co-occurrence network analysis, microbial physiologies matched the predominant geochemical conditions, and relatively stable community structures prevailed throughout the experiment. NO3- and NH4+ were more available to the microbes in the fluctuating regime’s alternating oxic/anoxic zone, leading to a more flexible network. The prevalent microbial metabolic pathways predicted using FAPROTAX were mainly associated with N and C metabolisms and significantly correlated with the taxonomic compositions. However, the dominant microbial groups were shared among all the soil samples. The limited spatial differences in microbial community structure under the fluctuating regime were likely due to a more pronounced role of microbial activity than microbial diversity, and similar drying-wetting conditions between the soil column and the filed site. Internally consistent geochemical and microbial depth stratifications were identified by (1) the redox potential and nitrogen species distributions, (2) spatial moment analysis of NO3- and NH4+ concentrations, (3) microbial community composition and (4) function. The five imposed hydrologic regimes generated distinct and coupled geochemical and microbial signatures. These signatures resulted from variations in the soil microbial community’s activity, rather than diversity or abundance.

Quantitative evaluation of particle–binder interactions in ceramic slurries via differential centrifugal sedimentation

In diverse materials science spanning from fine ceramics to lithium-ion batteries and fuel cells, the particle–binder interactions in slurries play a crucial role in governing the ultimate performance. Despite numerous efforts to date, quantitatively elucidating these hidden interactions has remained a longstanding challenge. Here, we demonstrate a dynamic approach to evaluate adsorptive interactions between ceramic particles and polymeric binders entangled in a slurry utilizing differential centrifugal sedimentation (DCS). Particles settling under a centrifugal force field impart significant viscous resistance on the adsorbed binder, leading to its detachment, influenced by particle size and density. This behaviour directly reflects the particle–binder interactions, and detailed DCS spectrum analysis enables the quantitative assessment of nano-Newton-order adsorption forces. An important finding is the strong correlation of these forces with the mechanical properties of the moulded products. Our results provide insight that forming a flexible network structure with appropriate interactions is essential for desirable formability.

Pure 3D Conducting Polymer Network for an Ultra-Sensitive and Flexible Hydrogen Gas Sensor at Room Temperature

Hydrogen (H2) is considered a next-generation energy source with diverse applications across valuable social, economic, and industrial sectors, including transportation, biomedical, power generation, and space exploration. Due to its low ignition energy and wide flammable range, there is an urgent global demand for ultrasensitive, lightweight, portable, wearable, and flexible hydrogen sensors to identify early gas leaks and prevent environmental and economic implications.In this study, we present a flexible hydrogen gas sensor based on conducting polymers, namely polypyrrole (PPY) with graphene nanoparticles, synthesized by the bi-continuous microemulsion polymerization method. The one-step scalable synthetic procedure enables the design of 1D, 2D, and 3D conducting polymers and their nanocomposites with tunable morphological and electrical properties. The unique multilayered oil-to-water nature of the bi-continuous microemulsion allows the fabrication of pure mesoporous 3D conducting polymers and their composites without the addition of any cross-linkers with excellent electrical properties. The 3D pure porous polymer network acts as a monolithic conducting framework that facilitates electron and ion transport and promotes the diffusion of molecules and ions more efficiently than 1D or 2D structures. The soft and porous nature of the polymer improves the interaction and adsorption of the target gas molecules and sensing material by increasing the number of active sites.This 3D PPY_graphene flexible network synthesized by the bi-continuous microemulsion method is successfully utilized to fabricate an ultra-sensitive flexible gas sensor. The performance of the sensor can be modified based on the reactor design. The sensor exhibits a notable response to 1 ppm of H2 gas. At room temperature, the fabricated sensor displayed 14 and 20 seconds of response and recovery time, respectively. To the best of our knowledge, no study has explored the application of 3D pure conducting polymers and their nanocomposites for hydrogen gas sensors using the bi-continuous microemulsion polymerization method, yielding such exceptional sensing results. Figure 1

Floating meta-bubbles: aerial gateway and routing on the sky

Reflecting intelligent surface technology (RIS) is regarded as a key enabler of the sixth-generation (6G) communication system. It provides the ability to reshape radio channels through passively reflecting beams in a reconstructive manner. Furthermore, aerial RIS (ARIS) introduces more flexibility in providing line-of-sight (LOS) links. Unfortunately, most of the related research efforts supposed the system as a planar RIS mounted on a satellite, unmanned aerial vehicle (UAV), or balloon despite reported limitations of planar RISs. The essential problem in designing any planar RIS network resides in mutual orientation and alignment difficulty, especially under random fluctuation of position/orientation due to wind conditions or UAV wobbling in the hover state. So, this paper highlights spherical RIS (bubble) as the optimal choice for aerial beam routing where the orientation/rotation can be completely relaxed. It outperforms planar RIS in terms of RIS networking flexibility, dead zone relaxation, and coverage extension. Consequently, due to the added degrees of freedom, many new deployment scenarios/use cases are recommended such as introducing meta-bubbles as intermediate gateways between satellite and ground nodes and extending network infrastructure installation down to the client level to enhance its visibility and throughput. Simulations demonstrate the superiority of meta-bubbles in minimizing channel loss over successive multi-hop routing.

Synthesis of flexible inter-plant heat exchanger networks: A decomposition method considering intermedium fluid circles

The traditional methods for synthesizing flexible heat exchanger networks (HENs) are not directly applicable to inter-plant HEN challenges, primarily due to the spread of system uncertainty across plants via intermedium fluid circles. This complicates the synthesis process significantly. To tackle this issue, this study proposes a decomposed stepwise methodology to facilitate the flexible synthesis of the inter-plant HENs performing indirect heat integration. A decomposition strategy is proposed to divide the overall network into manageable sub-networks by dissecting the intermedium fluid circles. To address the variability in intermedium fluid temperatures, a temperature fluctuation analysis approach is developed and a heuristic rule is introduced to maintain the temperature feasibility of the intermedium fluids. To ensure adequate flexibility and cost-effectiveness of the designed networks, flexibility analysis and network retrofit steps are conducted through model-based optimization techniques. The efficacy of the method is demonstrated through two case studies, showing its potential in achieving the desired operational flexibility for inter-plant HENs.

CrossHomo: Cross-Modality and Cross-Resolution Homography Estimation.

Multi-modal homography estimation aims to spatially align the images from different modalities, which is quite challenging since both the image content and resolution are variant across modalities. In this paper, we introduce a novel framework namely CrossHomo to tackle this challenging problem. Our framework is motivated by two interesting findings which demonstrate the mutual benefits between image super-resolution and homography estimation. Based on these findings, we design a flexible multi-level homography estimation network to align the multi-modal images in a coarse-to-fine manner. Each level is composed of a multi-modal image super-resolution (MISR) module to shrink the resolution gap between different modalities, followed by a multi-modal homography estimation (MHE) module to predict the homography matrix. To the best of our knowledge, CrossHomo is the first attempt to address the homography estimation problem with both modality and resolution discrepancy. Extensive experimental results show that our CrossHomo can achieve high registration accuracy on various multi-modal datasets with different resolution gaps. In addition, the network has high efficiency in terms of both model complexity and running speed.

Influence of monomer structure and catalyst concentration on topological transition and dynamic properties of dicarboxylic acid‐epoxy vitrimers

AbstractThis study delineates the dependence of thermophysical behavior of acid‐epoxy vitrimers on their formulation. The stress relaxation due to the bond exchange reaction and the glass transition temperature of acid epoxy vitrimers are investigated, with respect to the influence of catalyst content and acid chain length. This is carried out for a range of dicarboxylic acids and catalyst concentrations formulated and characterized using calorimetry and dynamic mechanical analysis. The influence of acid chain length on the bond exchange rate, topological transition, and glass transition temperatures of the vitrimers is found to be significant. The activation energy of the exchange reaction varies over a wide range from 73 to 104 kJ/mol and the topology freezing temperature from 66 to 136°C with the behavior governed by the interplay between crosslinking density, network flexibility and density and distance of functional groups, with an increase of catalyst concentration leading to lower topological transition temperature and the dependence on chain length showing non‐monotonic behavior. The glass transition decreases by about 30°C as the carbon chain length increases from 6 to 14 carbons due to enhanced monomer flexibility and is not affected by the concentration of catalyst.

High modularity, more flexible of brain networks in patients with mild to moderate motor impairments after stroke

Stroke is recognized as a network communication disorder. Advances in neuroimaging technologies have enhanced our comprehension of dynamic cerebral alterations. However, different levels of motor function impairment after stroke may have different patterns of brain reorganization. Abnormal and adaptive patterns of brain activity in mild-to-moderate motor function impairments after stroke remain still underexplored. We aim to identify dynamic patterns of network remodeling in stroke patients with mild-to-moderate impairment of motor function. fMRI data were obtained from 30 stroke patients and 31 healthy controls to establish a spatiotemporal multilayer modularity model. Then, graph-theoretic measures, including modularity, flexibility, cohesion, and disjointedness, were calculated to quantify dynamic reconfiguration. Our findings reveal that the post-stroke brain exhibited higher modular organization, as well as heightened disjointedness, compared to HCs. Moreover, analyzing from the network level, we found increased disjointedness and flexibility in the Default mode network (DMN), indicating that brain regions tend to switch more frequently and independently between communities and the dynamic changes were mainly driven by DMN. Notably, modified functional dynamics positively correlated with motor performance in patients with mild-to-moderate motor impairment. Collectively, our research uncovered patterns of dynamic community reconstruction in multilayer networks following stroke. Our findings may offer new insights into the complex reorganization of neural function in post-stroke brain.

Design of deep eutectic solvents for multiple perfluoroalkyl substances removal: Energy-based screening and mechanism elucidation

The current landscape of perfluoroalkyl substances (PFAS) extraction methodologies presents significant challenges, particularly for multiple PFAS with different carbon chain lengths. This study introduced an energy-driven strategic approach for screening deep eutectic solvents (DESs) to effectively remove a diverse range of PFAS, including perfluoroalkylcarboxylic acids (PFCAs), perfluoroalkanesulfonic acids (PFSAs), and perfluoroalkyl amides (FAAs), from contaminated environments (total 13 target compounds). Utilizing energy-based screening, we identified DES candidates with high affinity for a spectrum of PFAS compounds from 1234 potential starting materials of eutectic systems. Key findings revealed the superior removal efficiency of tributylphosphineoxide/2-methylpiperazine system, exceeding 99 % for various PFAS with different carbon chain lengths in real environmental water samples. Additionally, we elucidated the molecular interactions between DESs and PFAS through ab initio molecular dynamics (AIMD) simulations, providing valuable insights into the mechanisms governing the removal process. The mechanism of extraction involves hydrogen bond network topology and structural organization, with DESs capable of extracting PFAS while maintaining a weakly aggregated state of target molecules and minimizing the impact on the intrinsic structures of DES. The proposed system forms a dynamic, complementary, and flexible non-covalent interaction network structure with PFAS. The study advances the understanding of DES as a designable, effective, and sustainable alternative to conventional solvents for PFAS remediation, offering a significant contribution to environmental chemistry and green technology.

Bioinspired Low-Angle-Dependent Photonic Crystal Elastomer for Highly Sensitive Visual Strain Sensor.

Photonic crystals (PCs) possess unique photonic band gap properties that can be used in the field of sensors and smart displays if modulated on the micronano structure. Both nonclose-packed (NCP) structure and high refractive index (RI) contrast of PC play important roles in response sensitivity during stretching. Herein, we constructed an NCP-structured PC strain sensor with high RI by a novel coating-etching strategy. Stretch-induced changes in structural color correspond to the strength of the force, enabling the detection of the strength of the acting force by the naked eye. The flexible 3D cross-linked network constructed by poly(ethylene glycol) phenyl ether acrylate and pentaerythritol tetrakis(3-mercaptopropionate) endows the sensor with excellent elasticity and robustness. The designed PC strain sensor achieves high mechanochromic sensitivity (∼8.3 nm/%, 0.02 to 4.21 MPa) and a substantial reflection peak shift (Δλ = 249 nm). More importantly, the high RI contrast (Δn = 0.43) between CdS and polymers imparts isotropic optical properties, ensuring a broad viewing angle while avoiding misleading signals. The research provides a novel fabrication strategy to construct sensitive PC strain sensors, expanding the prospective applicability to human movement monitoring and secure message encryption.

Structured flexibility in recurrent neural networks via neuromodulation.

The goal of theoretical neuroscience is to develop models that help us better understand biological intelligence. Such models range broadly in complexity and biological detail. For example, task-optimized recurrent neural networks (RNNs) have generated hypotheses about how the brain may perform various computations, but these models typically assume a fixed weight matrix representing the synaptic connectivity between neurons. From decades of neuroscience research, we know that synaptic weights are constantly changing, controlled in part by chemicals such as neuromodulators. In this work we explore the computational implications of synaptic gain scaling, a form of neuromodulation, using task-optimized low-rank RNNs. In our neuromodulated RNN (NM-RNN) model, a neuromodulatory subnetwork outputs a low-dimensional neuromodulatory signal that dynamically scales the low-rank recurrent weights of an output-generating RNN. In empirical experiments, we find that the structured flexibility in the NM-RNN allows it to both train and generalize with a higher degree of accuracy than low-rank RNNs on a set of canonical tasks. Additionally, via theoretical analyses we show how neuromodulatory gain scaling endows networks with gating mechanisms commonly found in artificial RNNs. We end by analyzing the low-rank dynamics of trained NM-RNNs, to show how task computations are distributed.