Input-output Mapping Research Articles

An intelligent measurement method of the resonant frequency of ultrasonic scalpel transducers based on PSO-BP neural network

In the laboratory, using traditional methods to measure the resonant frequency of transducers is inefficient and costly. For this, in this study, the feature variables determining the resonant frequency of transducer was firstly analyzed with the equivalent circuit method. Based on the input–output mapping function of back propagation (BP) neural network, the nonlinear relationship model between feature variables and the resonant frequency was established. Then, particle swarm optimization (PSO) was used to optimize the initial weight and threshold of BP neural network to obtain the best parameter combination, thereby improving the output accuracy of BP neural network. The experimental data showed that it is effective to measure the resonant frequency of the transducer by PSO-BP neural network, and the measurement error was 1 Hz. Compared to traditional measurement method, the intelligent method has the advantages of low cost, high efficiency, fast response speed, and strong robustness.

Vehicle Motion State Prediction Method Integrating Point Cloud Time Series Multiview Features and Multitarget Interactive Information

A vehicle motion state prediction algorithm integrating point cloud timing multiview features and multitarget interaction information is proposed in this work to effectively predict the motion states of traffic participants around intelligent vehicles in complex scenes. The algorithm analyzes the characteristics of object motion that are affected by the surrounding environment and the interaction of nearby objects and is based on the complex traffic environment perception dual multiline light detection and ranging (LiDAR) technology. The time sequence aerial view map and time sequence front view depth map are obtained using real-time point cloud information perceived by the LiDAR. Time sequence high-level abstract combination features in the multiview scene are then extracted by an improved VGG19 network model and are fused with the potential spatiotemporal interaction of the multitarget operation state data extraction features detected by the laser radar by using a one-dimensional convolution neural network. A temporal feature vector is constructed as the input data of the bidirectional long-term and short-term memory (BiLSTM) network, and the desired input-output mapping relationship is trained to predict the motion state of traffic participants. According to the test results, the proposed BiLSTM model based on point cloud multiview and vehicle interaction information is better than other methods in predicting the state of target vehicles. The results can provide support for the research to evaluate the risk of intelligent vehicle operation environment.

Building Marginal Pattern Library With Unbiased Training Dataset for Enhancing Model-Free Load-ED Mapping

Input-output mapping for a given power system problem, such as loads versus economic dispatch (ED) results, has been demonstrated to be learnable through artificial intelligence (AI) techniques, including neural networks. However, the process of identifying and constructing a comprehensive dataset for the training of such input-output mapping remains a challenge to be solved. Conventionally, load samples are generated by a pre-defined distribution, and then ED is solved based on those load samples to form training datasets, but this paper demonstrates that such dataset generation is biased regarding load-ED mapping. The marginal unit and line congestion (i.e., marginal pattern) exhibit a unique characteristic called “step change” in which the marginal pattern changes when the load goes from one critical loading level (CLL) to another, and there is no change of marginal units within the interval of the two adjacent CLLs. Those loading intervals differ significantly in size. The randomly generated training dataset overfills intervals with large sizes and underfits intervals with small sizes, so it is biased. In this paper, three algorithms are proposed to construct a marginal pattern library to examine this bias according to different computational needs, and an enhancement algorithm is proposed to eliminate the bias in the load-ED training dataset generation. Three illustrative test cases demonstrate the proposed algorithms, and comparative studies are constructed to show the superiority of the enhanced, unbiased training dataset.

Low-Gain Stability of Projected Integral Control for Input-Constrained Discrete-Time Nonlinear Systems

We consider the problem of zeroing an error output of a nonlinear discrete-time system in the presence of constant exogenous disturbances, subject to hard convex constraints on the input signal. The design specification is formulated as a variational inequality, and we adapt a forward-backward splitting algorithm to act as an integral controller which ensures that the input constraints are met at each time step. We establish a low-gain stability result for the closed-loop system when the plant is exponentially stable, generalizing previously known results for integral control of discrete-time systems. Specifically, it is shown that if the composition of the plant equilibrium input-output map and the integral feedback gain is strongly monotone, then the closed-loop system is exponentially stable for all sufficiently small integral gains. The method is illustrated via application to a four-tank process.

Game-Theoretic Feedback-Based Optimization

This paper examines the intersection between feedback-based optimization problems and distributed Nash equilibrium seeking algorithms. We consider a modification of typical GNE-seeking problems with affine coupling constraints, wherein each agent's objective additionally depends on the measurable output of a nonlinear input-output mapping. Operator-theoretic methods are leveraged to develop an online distributed algorithm for this class of problems, with convergence criteria provided. We illustrate the algorithm via an application to coordination of distributed energy resources in a power distribution feeder.

A Double-Layer Multi-Resolution Classification Model for Decoding Spatiotemporal Patterns of Spikes With Small Sample Size.

We build a double-layer, multiple temporal-resolution classification model for decoding single-trial spatiotemporal patterns of spikes. The model takes spiking activities as input signals and binary behavioral or cognitive variables as output signals and represents the input-output mapping with a double-layer ensemble classifier. In the first layer, to solve the underdetermined problem caused by the small sample size and the very high dimensionality of input signals, B-spline functional expansion and L1-regularized logistic classifiers are used to reduce dimensionality and yield sparse model estimations. A wide range of temporal resolutions of neural features is included by using a large number of classifiers with different numbers of B-spline knots. Each classifier serves as a base learner to classify spatiotemporal patterns into the probability of the output label with a single temporal resolution. A bootstrap aggregating strategy is used to reduce the estimation variances of these classifiers. In the second layer, another L1-regularized logistic classifier takes outputs of first-layer classifiers as inputs to generate the final output predictions. This classifier serves as a meta-learner that fuses multiple temporal resolutions to classify spatiotemporal patterns of spikes into binary output labels. We test this decoding model with both synthetic and experimental data recorded from rats and human subjects performing memory-dependent behavioral tasks. Results show that this method can effectively avoid overfitting and yield accurate prediction of output labels with small sample size. The double-layer, multi-resolution classifier consistently outperforms the best single-layer, single-resolution classifier by extracting and utilizing multi-resolution spatiotemporal features of spike patterns in the classification.

Feedback-Based Optimization With Sub-Weibull Gradient Errors and Intermittent Updates

This letter considers a feedback-based projected gradient method for optimizing systems modeled as algebraic maps. The focus is on a setup where the gradient is corrupted by random errors that follow a sub-Weibull distribution, and where the measurements of the output – which replace the input-output map of the system in the algorithmic updates – may not be available at each iteration. The sub-Weibull error model is particularly well-suited in frameworks where the cost of the problem is learned via Gaussian Process (GP) regression (from functional evaluations) concurrently with the execution of the algorithm; however, it also naturally models setups where nonparametric methods and neural networks are utilized to estimate the cost. Using the sub-Weibull model, and with Bernoulli random variables modeling missing measurements of the system output, we show that the online algorithm generates points that are within a bounded error from the optimal solutions. In particular, we provide error bounds in expectation and in high probability. Numerical results are presented in the context of a demand response problem in smart power grids.

MIMO auto-regressive modeling-based generalized predictive control for grid-connected hybrid systems

In this paper, we propose an energy management model for grid-connected hybrid systems. This model is based on the coupling of a multiple input multiple output (MIMO) AutoRegressive Moving Average with eXogenous inputs (ARMAX) model with generalized predictive control (GPC). The ARMAX model performs input–output mapping of the electrical system data in terms of energy costs and power values. The GPC system receives the input–output mapping provided by the ARMAX model and performs energy management by minimizing the consumed energy cost in the load. Cost minimization is accomplished by optimizing the use of renewable energy sources, thus consuming less energy from the grid. The effectiveness of the proposed model is verified, comparing its computational and experimental results to those of a conventional control model and a model predictive control approach presented in the literature.

Systematic Input-Output Mapping Reveals Structural Plasticity of VTA Dopamine Neurons-Zona Incerta Loop Underlying the Social Buffering Effects in Learned Helplessness.

A common phenomenon called social buffering (SB), communication within conspecific animals is a benefit for a stressed individual to better recover from aversive events, is crucial to all mammals. Although the dopamine reward system has been implicated in SB, it is not clear which neuronal populations are relevant and how they contribute. Here, we adopted a learned helplessness (LH) animal model of depression and found that LH subjects housed with a conspecific partner show better performance in the shuttle box test, showing that SB improves the stress-coping abilities to deal with stress. Bidirectional manipulation of ventral tegmental area (VTA) dopamine neurons by chemogenetic tools can mimic or block the SB effect in LH mice. To screen for SB-induced structure plasticity of VTA dopamine neurons, we employed viral genetic tools for mapping input and output architecture and found LH- and SB-triggered circuit-level changes in neuronal ensembles. Zona incerta (ZI), an overlapping brain region, was significantly changed in both anterograde and retrograde tracing during LH and SB. These results reveal a neural loop with structural plasticity between VTA dopamine neurons and ZI underlies the SB effects in LH and lays a foundation for studying how VTA dopamine neurons regulate SB-related neural circuits.

Analytical evidence of nonlinearity in qubits and continuous-variable quantum reservoir computing

The natural dynamics of complex networks can be harnessed for information processing purposes. A paradigmatic example are artificial neural networks used for machine learning. In this context, quantum reservoir computing (QRC) constitutes a natural extension of the use of classical recurrent neural networks using quantum resources for temporal information processing. Here, we explore the fundamental properties of QRC systems based on qubits and continuous variables. We provide analytical results that illustrate how nonlinearity enters the input–output map in these QRC implementations. We find that the input encoding through state initialization can serve to control the type of nonlinearity as well as the dependence on the history of the input sequences to be processed.

Global Self-Optimizing Control With Data-Driven Optimal Selection of Controlled Variables With Application to Chiller Plant

AbstractIn this paper, we propose a global self-optimizing control (SOC) approach, where nonlinear dynamic model is obtained from historical data of plant operation via the framework of sparse identification of nonlinear dynamics (SINDy) combined with regularized regression. With the nonlinear static input–output map obtained by forcing steady-state operation, the globally optimal solutions of controlled variables (CVs) can be found by tracking the necessary conditions of optimality (NCO) in an analytical fashion. After validation with a numerical example, the proposed method is evaluated using a modelica-based dynamic model of a chilled-water plant. The economic objective for chiller plant operation is to minimize the total power of compressor, condenser water pump, and cooling tower fan, while the cooling tower fan speed and condenser water mass flow rate are used as manipulated inputs. The operating data are generated based on realistic ambient and load conditions and a best-practice rule-based strategy for chiller operation. The control structure with the SOC method yields a total power consumption close to the global optimum and substantially smaller than that of a best-practice rule-based chiller plant control strategy. The proposed method promises a global SOC solution using dynamic operation data, for cost-effective and adaptive control structure optimization.

Nanoscale neural network using non-linear spin-wave interference

We demonstrate the design of a neural network hardware, where all neuromorphic computing functions, including signal routing and nonlinear activation are performed by spin-wave propagation and interference. Weights and interconnections of the network are realized by a magnetic-field pattern that is applied on the spin-wave propagating substrate and scatters the spin waves. The interference of the scattered waves creates a mapping between the wave sources and detectors. Training the neural network is equivalent to finding the field pattern that realizes the desired input-output mapping. A custom-built micromagnetic solver, based on the Pytorch machine learning framework, is used to inverse-design the scatterer. We show that the behavior of spin waves transitions from linear to nonlinear interference at high intensities and that its computational power greatly increases in the nonlinear regime. We envision small-scale, compact and low-power neural networks that perform their entire function in the spin-wave domain.

Decentralized uniform finite‐time stabilization of large‐scale nonlinear networks by small‐gain theorems

AbstractWe solve the problem of decentralized uniform finite‐time stabilization for a large‐scale network of nonlinear systems in triangular form which are not feedback linearizable and whose input–output maps are not invertible. For this, we provide a new recursive design to satisfy the conditions of a certain modification of the small‐gain theorems for the case of the uniform finite‐time stability of large‐scale networks. In the general case, we consider the general triangular form systems whose input–output links are surjections only. In addition, we consider the special case of the triangular form systems with polynomial integrators, which is also needed as the first step of the proof of our main result. In the latter case, the design becomes constructive, the decentralized stabilizers can be designed explicitly, and this is demonstrated by examples.

Efficient and transferable machine learning potentials for the simulation of crystal defects in bcc Fe and W

Data-driven, or machine learning (ML), approaches have become viable alternatives to semiempirical methods to construct interatomic potentials, due to their capacity to accurately interpolate and extrapolate from first-principles simulations if the training database and descriptor representation of atomic structures are carefully chosen. Here, we present highly accurate interatomic potentials suitable for the study of dislocations, point defects, and their clusters in bcc iron and tungsten, constructed using a linear or quadratic input-output mapping from descriptor space. The proposed quadratic formulation, called quadratic noise ML, differs from previous approaches, being strongly preconditioned by the linear solution. The developed potentials are compared to a wide range of existing ML and semiempirical potentials, and are shown to have sufficient accuracy to distinguish changes in the exchange-correlation functional or pseudopotential in the underlying reference data, while retaining excellent transferability. The flexibility of the underlying approach is able to target properties almost unattainable by traditional methods, such as the negative divacancy binding energy in W or the shape and the magnitude of the Peierls barrier of the $\frac{1}{2}\ensuremath{\langle}111\ensuremath{\rangle}$ screw dislocation in both metals. We also show how the developed potentials can be used to target important observables that require large time-and-space scales unattainable with first-principles methods, though we emphasize the importance of thoughtful database design and degrees of nonlinearity of the descriptor space to achieve the appropriate passage of information to large-scale calculations. As a demonstration, we perform direct atomistic calculations of the relative stability of $\frac{1}{2}\ensuremath{\langle}111\ensuremath{\rangle}$ dislocations loops and three-dimensional C15 clusters in Fe and find the crossover between the formation energies of the two classes of interstitial defects occurs at around 40 self-interstitial atoms. We also compute the kink-pair formation energy of the $\frac{1}{2}\ensuremath{\langle}111\ensuremath{\rangle}$ screw dislocation in Fe and W, finding good agreement with density functional theory informed line tension models that indirectly measure those quantities. Finally, we exploit the excellent finite-temperature properties to compute vacancy formation free energies with full anharmonicity in thermal vibrations. The presented potentials thus open up many avenues for systematic investigation of free-energy landscape of defects with ab initio accuracy.

A multimodal cell census and atlas of the mammalian primary motor cortex

Here we report the generation of a multimodal cell census and atlas of the mammalian primary motor cortex as the initial product of the BRAIN Initiative Cell Census Network (BICCN). This was achieved by coordinated large-scale analyses of single-cell transcriptomes, chromatin accessibility, DNA methylomes, spatially resolved single-cell transcriptomes, morphological and electrophysiological properties and cellular resolution input–output mapping, integrated through cross-modal computational analysis. Our results advance the collective knowledge and understanding of brain cell-type organization1–5. First, our study reveals a unified molecular genetic landscape of cortical cell types that integrates their transcriptome, open chromatin and DNA methylation maps. Second, cross-species analysis achieves a consensus taxonomy of transcriptomic types and their hierarchical organization that is conserved from mouse to marmoset and human. Third, in situ single-cell transcriptomics provides a spatially resolved cell-type atlas of the motor cortex. Fourth, cross-modal analysis provides compelling evidence for the transcriptomic, epigenomic and gene regulatory basis of neuronal phenotypes such as their physiological and anatomical properties, demonstrating the biological validity and genomic underpinning of neuron types. We further present an extensive genetic toolset for targeting glutamatergic neuron types towards linking their molecular and developmental identity to their circuit function. Together, our results establish a unifying and mechanistic framework of neuronal cell-type organization that integrates multi-layered molecular genetic and spatial information with multi-faceted phenotypic properties.

Fault detection and isolation of actuator failures in jet engines using adaptive dynamic programming

This paper presents a adaptive dynamic programming-based fault detection and isolation (FDI) scheme to detect and isolate faults in an aircraft jet engine. To this end, the weights in Actor-Critic neural networks are first tuned to learn the input-output map of the jet engine considering its multiple working modes. The convergences of the trainings in Critic-Actor neural networks are strictly proved without knowing the drift dynamics and the input dynamics in the presence of unknown nonlinearities and approximation errors. Using the residuals that are generated by measuring the difference of each network output and the measured engine output, various criteria are established for accomplishing the fault diagnosis task, that addresses the problem of fault detection and isolation of the system components. A number of simulation studies are carried out for combustion chamber of a single-spool jet engine to demonstrate and illustrate the advantages, capabilities, and performance of our proposed fault diagnosis scheme.

POD-Enhanced Deep Learning-Based Reduced Order Models for the Real-Time Simulation of Cardiac Electrophysiology in the Left Atrium

The numerical simulation of multiple scenarios easily becomes computationally prohibitive for cardiac electrophysiology (EP) problems if relying on usual high-fidelity, full order models (FOMs). Likewise, the use of traditional reduced order models (ROMs) for parametrized PDEs to speed up the solution of the aforementioned problems can be problematic. This is primarily due to the strong variability characterizing the solution set and to the nonlinear nature of the input-output maps that we intend to reconstruct numerically. To enhance ROM efficiency, we proposed a new generation of non-intrusive, nonlinear ROMs, based on deep learning (DL) algorithms, such as convolutional, feedforward, and autoencoder neural networks. In the proposed DL-ROM, both the nonlinear solution manifold and the nonlinear reduced dynamics used to model the system evolution on that manifold can be learnt in a non-intrusive way thanks to DL algorithms trained on a set of FOM snapshots. DL-ROMs were shown to be able to accurately capture complex front propagation processes, both in physiological and pathological cardiac EP, very rapidly once neural networks were trained, however, at the expense of huge training costs. In this study, we show that performing a prior dimensionality reduction on FOM snapshots through randomized proper orthogonal decomposition (POD) enables to speed up training times and to decrease networks complexity. Accuracy and efficiency of this strategy, which we refer to as POD-DL-ROM, are assessed in the context of cardiac EP on an idealized left atrium (LA) geometry and considering snapshots arising from a NURBS (non-uniform rational B-splines)-based isogeometric analysis (IGA) discretization. Once the ROMs have been trained, POD-DL-ROMs can efficiently solve both physiological and pathological cardiac EP problems, for any new scenario, in real-time, even in extremely challenging contexts such as those featuring circuit re-entries, that are among the factors triggering cardiac arrhythmias.

Phase retrieval based on difference map and deep neural networks

Phase retrieval occurs in many research areas. There are some classical phase retrieval methods such as hybrid input-output (HIO) and difference map (DM). However, phase retrieval results are sensitive to noise, and the reconstructed images always include artefacts. In this paper, we use the DM algorithm together with DNN to get better phase retrieval results. We train one deep neural network using amplitude images and phase images, respectively. First, using DM, we get initial reconstructed amplitude and phase results. Then, using DNN improves both amplitude and phase results. Finally, using the DM algorithm again improves the DNN results further. The numerical experimental results show that using DM gives better results than HIO, and using DNN improves phase information better than just using DNN to train for amplitude information alone. Compared with only using DNN improves amplitude methods, our method using DM plus DNN plus DM yields a better reconstruction performance for both amplitude and phase.

Control on the Manifolds of Mappings with a View to the Deep Learning

Deep learning of the artificial neural networks (ANN) can be treated as a particular class of interpolation problems. The goal is to find a neural network whose input-output map approximates well the desired map on a finite or an infinite training set. Our idea consists of taking as an approximant the input-output map, which arises from a nonlinear continuous-time control system. In the limit such control system can be seen as a network with a continuum of layers, each one labelled by the time variable. The values of the controls at each instant of time are the parameters of the layer.

An intelligent parameter determination approach in iterative learning control

This paper introduces an approach for determining parameters of learning function in iterative learning control without using the mathematical model of the controlled plant, which guarantees definitely the convergence requirement of the learning process. The learning parameter is intelligently computed based on system’s tracking errors from the two previous consecutive trials by minimizing a suitable cost function. Then, analysis of stability and convergence using injective property and Jacobian matrix of the input-output mapping is also given in this paper. Finally, some numerical simulations with actual system model of robot manipulators are performed to illustrate and support the proposed method.