Input Dimension Research Articles

An adaptive dynamic surface control strategy for suppressing pathological oscillations in a neural mass model

Pathological oscillations that have a frequency in the [Formula: see text] band are widely recognized to be involved in Parkinson’s disease. Now, we present an adaptive dynamic surface control strategy for suppressing pathological oscillations that exist in the Parkinsonian state. First, the interactions between the subthalamic nucleus and external globus pallidus are fully considered and establish a neural mass model of Parkinson’s disease. Next, by reasonable state transformation, suppressing pathological oscillations can be converted into a tracking control study of a pure-feedback nonlinear system. Moreover, the dynamic surface control technique adopted reduces the dimensionality of the neural network input and effectively eliminates the complexity explosion problem commonly associated with existing methods. By Lyapunov stability analysis, it can be obtained that all the signals of the resulting closed-loop system are bounded, and the tracking error converges to a small neighborhood of zero. Finally, the simulation results demonstrate the effectiveness of the designed control strategy. This work may provide an effective approach to closed-loop deep brain stimulation optimization for the alleviation of the Parkinsonian state.

What can be learnt with wide convolutional neural networks?*

Abstract Understanding how convolutional neural networks (CNNs) can efficiently learn high-dimensional functions remains a fundamental challenge. A popular belief is that these models harness the local and hierarchical structure of natural data such as images. Yet, we lack a quantitative understanding of how such structure affects performance, for example the rate of decay of the generalisation error with the number of training samples. In this paper, we study infinitely wide deep CNNs in the kernel regime. First, we show that the spectrum of the corresponding kernel inherits the hierarchical structure of the network, and we characterise its asymptotics. Then, we use this result together with generalisation bounds to prove that deep CNNs adapt to the spatial scale of the target function. In particular, we find that if the target function depends on low-dimensional subsets of adjacent input variables then the decay of the error is controlled by the effective dimensionality of these subsets. Conversely, if the target function depends on the full set of input variables then the error decay is controlled by the input dimension. We conclude by computing the generalisation error of a deep CNN trained on the output of another deep CNN with randomly initialised parameters. Interestingly, we find that, despite their hierarchical structure, the functions generated by infinitely wide deep CNNs are too rich to be efficiently learnable in high dimensions.

High-dimensional asymptotics of denoising autoencoders*

Abstract We address the problem of denoising data from a Gaussian mixture using a two-layer non-linear autoencoder with tied weights and a skip connection. We consider the high-dimensional limit where the number of training samples and the input dimension jointly tend to infinity while the number of hidden units remains bounded. We provide closed-form expressions for the denoising mean-squared test error. Building on this result, we quantitatively characterize the advantage of the considered architecture over the autoencoder without the skip connection that relates closely to principal component analysis. We further show that our results accurately capture the learning curves on a range of real data sets.

Strata Design for Variance Reduction in Stochastic Simulation

Stratified sampling is one of the powerful variance reduction methods for analyzing system performance, such as reliability, with stochastic simulation. It divides the input space into disjoint subsets, called strata, to draw samples from each stratum. Partitioning the input space properly and allocating greater computational effort to crucial strata can help accurately estimate system performance with a limited computational budget. How to create strata, however, has yet to be thoroughly examined. Strata design faces the curse of dimensionality and data scarcity as the input dimension increases. We analytically derive the optimal stratification structure that minimizes the estimation variance for univariate problems. Further, reconciling the optimal stratification into decision trees, we devise a robust algorithm for multi-dimensional problems. Numerical experiments and a wind turbine case study demonstrate the superiority of the proposed method in terms of variance reduction, leading to computational efficiency and scalability.

A high-precision data-driven radar modeling method for autonomous driving simulations testing

Autonomous driving represents the future of transportation, and simulation testing is a critical technology for ensuring the safety and reliability of autonomous driving. In simulation systems, a real-time and accurate radar model is crucial for enhancing the confidence of autonomous driving simulation systems. Existing radar models are typically simplistic and cannot accurately reflect the detection results of real radars. This paper introduces a data-driven radar modeling method that analyzes radar detection mechanisms to identify the model’s input parameters. Furthermore, this method decouples the model’s output parameters based on the differing mechanisms of radar output parameters and tailors model structures to align with the mechanistic characteristics of each parameter. Subsequently, considering the varied influences among input parameters, a segmented model structure is proposed that reduces input dimensionality while retaining critical information. Lastly, the model outputs are optimized using Kalman filtering. This study collects radar data from real vehicles to train the radar model. Testing has demonstrated that the radar model developed by this paper’s method yields outputs very close to actual radar predictions, with a significant improvement in prediction accuracy compared to traditional radar models. The model operates in approximately 0.8 ms, markedly faster than the real radar’s 72-millisecond detection cycle. This demonstrates that the radar model developed using this modeling method can predict radar detection results accurately and in real-time.

GNSS/IMU integrated navigation heading angle enhancement algorithm based on dual-antenna TDCP

In the integrated navigation system of Global Navigation Satellite System (GNSS) and Inertial Measurement Unit (IMU), attitude estimation, especially accurate estimation of heading angle, is particularly important for real-time monitoring of vehicle driving status. However, since IMU is divergent in the altitude channel, its error will gradually accumulate if it is not accurately constrained. Therefore, the estimation accuracy of heading angle is insufficient in vehicle-mounted applications where the heading angle changes frequently. In order to solve the problem of poor navigation attitude estimation accuracy in the loose combination of traditional GNSS and IMU, this paper proposes a GNSS/IMU integrated navigation heading angle enhancement algorithm based on dual-antenna epoch carrier phase difference (TDCP). This method solves the vehicle heading angle through dual-antenna TDCP to increase the input dimension of the integrated navigation filter fusion observation value, and uses Hatch filtering and anti-error adaptive filtering to improve the observation domain pseudorange accuracy and GNSS/IMU integrated navigation positioning and attitude determination performance respectively. Experimental results show that compared with the traditional GNSS/IMU integrated navigation method, the proposed algorithm improves the positioning and speed measurement accuracy in the three-dimensional direction by and respectively 22.12%, 41.27%and the vehicle heading angle accuracy is improved 46.29%.

A Sequential Stein's Method for Faster Training of Additive Index Models.

The additive index models (AIMs) can be viewed as a kind of artificial neural networks based on nonparametric activation or so-called ridge functions. Recently, they are shown to achieve enhanced explainability after incorporating various interpretability constraints. However, the training of AIMs by either the backfitting algorithm or the joint stochastic optimization is known to be very slow for especially high dimensional inputs. In this article, we propose a novel sequential approach based on the celebrated Stein's lemma. The proposed SeqStein method can successfully decouple the training of AIMs into two separable steps, namely, the following: 1) Stein's estimation of the projection indices and 2) nonparametric estimation of ridge functions using the smoothing splines. We show through numerical experiments that the SeqStein algorithm is not only more efficient for training AIMs, but also inclined to produce more interpretable models that have smooth ridge functions with sparse and nearly orthogonal projection indices.

Predictive modeling using Copula Particle Filter and Adaptive Network-Based Fuzzy Inference

This paper introduces a novel prediction algorithm, CPF-ANFIS, designed to overcome the challenges posed by high-dimensional input data in Adaptive Neuro-Fuzzy Inference Systems (ANFIS). ANFIS's performance deteriorates with increasing input dimensionality due to the distortion of its membership functions. To address this limitation, CPF-ANFIS leverages a two-stage approach: A Copula Particle Filter (CPF) for robust state estimation and ANFIS for nonlinear mapping. By incorporating copulas, CPF effectively addresses the impoverishment and degeneracy problems commonly encountered in traditional particle filters. This enhanced robustness allows for more accurate state estimation, which in turn improves the overall performance of the CPF-ANFIS algorithm. By decoupling state estimation from nonlinear modeling, CPF-ANFIS effectively mitigates the curse of dimensionality. The proposed method is evaluated on real-world applications, such as hybrid PV-wind systems and SLAM. Experimental results demonstrate that CPF-ANFIS consistently outperforms ANFIS and the Copula Particle Filter individually, as well as previously proposed methods such as ANFIS-PF, highlighting its effectiveness in achieving accurate predictions under challenging conditions. The results show that the CPF-ANFIS algorithm increases prediction accuracy by at least 5% compared to using each algorithm separately.

A novel directional simulation method for estimating failure possibility

Failure possibility plays a crucial role in assessing the safety level of structures under fuzzy uncertainty. However, the traditional fuzzy simulation method suffers from computational inefficiency as it requires a large number of samples for accurate estimation. To address this issue, a directional simulation method is proposed to improve the efficiency of estimating failure possibility. The directional simulation method reformulates the failure possibility estimation into two key steps: the generation of direction samples and the estimation of conditional failure possibility under each direction sample in the polar coordinate system of the standard fuzzy space. To ensure direction uniformity, these direction samples are generated by adopting a good lattice point set based on stratified sampling on the unit hypercube. The conditional failure possibility under each direction sample is estimated by solving the minimum root of a nonlinear equation. The proposed method effectively reduces the dimensionality of the fuzzy input and greatly improves the computational efficiency. To further enhance efficiency, an adaptive Kriging model is embedded into the directional simulation method to reduce the number of performance function evaluations. Four examples are performed to illustrate the accuracy and efficiency of the proposed method. The results highlight the superiority of the directional simulation method over the traditional fuzzy simulation method, offering substantial improvements in computational efficiency while maintaining high estimation accuracy.

Data-driven state-of-charge estimation of a lithium-ion battery pack in electric vehicles based on real-world driving data

State-of-charge (SOC) is one of the most important indicators in the battery management system (BMS) of electric vehicles (EVs). Accurate online estimation of battery pack SOC is essential for ensuring safety by preventing overcharging and overdischarging, as well as for facilitating proper operation and efficient energy utilization. This paper proposes a deep learning model, CNN-BiLSTM-Attention, to drive the battery pack SOC estimation by datasets collected from actual operating EVs. Multivariate time series (MVTS) signals such as vehicle speed, battery voltage, current and temperature are carefully selected as inputs to the model. Additionally, a three-parameter Weibull probability model is used to model the terminal voltage distribution of all cells within the investigated battery pack, and the evolution of cell terminal voltage inconsistency concerning depth of discharge (DOD) is described from the perspective of dispersion and symmetry. By adding these distribution features to the input dimension of the CNN-BiLSTM-Attention model, the estimation accuracy of battery pack SOC has been further improved. Experimental results indicate that when the window length (WL) is set to 25 sampling time steps and the window overlap rate is >72 %, the SOC estimation error is <3 % and the root mean square error (RMSE) is <1.12 % within the maximum DOD variation range experienced by the battery pack. Especially when the battery pack is in the discharge platform period of 30 %–100 % SOC, the SOC estimation error is reduced to <2 %. After adding the Weibull distribution features that reflect the cell inconsistency evolution into the model's input dimensions, the overall SOC estimation RMSE is reduced by about 31 %, especially by about 51 % at lower SOC levels (below 31 %).

Systematic changes in neural selectivity reflect the acquired salience of category-diagnostic dimensions.

Humans and other animals develop remarkable behavioral specializations for identifying, differentiating, and acting on classes of ecologically important signals. Ultimately, this expertise is flexible enough to support diverse perceptual judgments: a voice, for example, simultaneously conveys what a talker says as well as myriad cues about her identity and state. Mature perception across complex signals thus involves both discovering and learning regularities that best inform diverse perceptual judgments, and weighting this information flexibly as task demands change. Here, we test whether this flexibility may involve endogenous attentional gain to task-relevant dimensions. We use two prospective auditory category learning tasks to relate a complex, entirely novel soundscape to four classes of "alien identity" and two classes of "alien size." Identity, but not size, categorization requires discovery and learning of patterned acoustic input situated in one of two simultaneous, frequency-delimited bands. This allows us to capitalize on the coarsely segregated frequency-band-specific channels of auditory tonotopic maps using fMRI to ask whether category-relevant perceptual information is prioritized relative to simultaneous, uninformative information. Among participants expert at alien identity categorization, we observe prioritization of the diagnostic frequency band that persists even when the diagnostic information becomes irrelevant in the size categorization task. Tellingly, the neural selectivity evoked implicitly in categorization aligns closely with activation driven by explicit, sustained selective attention to other sounds presented in the same frequency band. Additionally, we observe fingerprints of individual differences in the learning trajectories taken to achieve expert-level categorization in patterns of neural activity associated with the diagnostic dimension. In all, this indicates that acquiring categories can drive the emergence of acquired attentional salience to dimensions of acoustic input.

Can a Hebbian-like learning rule be avoiding the curse of dimensionality in sparse distributed data?

It is generally assumed that the brain uses something akin to sparse distributed representations. These representations, however, are high-dimensional and consequently they affect classification performance of traditional Machine Learning models due to the "curse of dimensionality". In tasks for which there is a vast amount of labeled data, Deep Networks seem to solve this issue with many layers and a non-Hebbian backpropagation algorithm. The brain, however, seems to be able to solve the problem with few layers. In this work, we hypothesize that this happens by using Hebbian learning. Actually, the Hebbian-like learning rule of Restricted Boltzmann Machines learns the input patterns asymmetrically. It exclusively learns the correlation between non-zero values and ignores the zeros, which represent the vast majority of the input dimensionality. By ignoring the zeros the "curse of dimensionality" problem can be avoided. To test our hypothesis, we generated several sparse datasets and compared the performance of a Restricted Boltzmann Machine classifier with some Backprop-trained networks. The experiments using these codes confirm our initial intuition as the Restricted Boltzmann Machine shows a good generalization performance, while the Neural Networks trained with the backpropagation algorithm overfit the training data.

Approach angle estimation method for ships based on deep learning

Accurately estimating the ship's approach angle is crucial during berthing, or passing through navigational structures. It forms the foundation for ship navigation decisions. This study introduces the LiDAR point cloud-based ship pose perception network (PP-Net), a novel deep learning-based method for acquiring ship approach angles. Addressing the challenge of manually annotating approach angles, the ship pose perception method based on point cloud feature distribution (FD) was proposed to obtain initial angle estimates. The max-pooling function was utilized to address the disorder of point clouds, and a multi-resolution feature extractor was employed to enhance the network's robustness in extracting features from inconsistently dense point clouds. By using iterative farthest point sampling (IFPS), datasets with varying numbers of point clouds were standardized to a fixed number of points, thereby normalizing input dimensions and improving computational efficiency. Additionally, a novel loss function was proposed to improve the model's predictive accuracy. Experimental validation conducted in the scenario of ships entering ship lifts demonstrates the effectiveness of the proposed method. The outcomes can support safe navigation for ships in restricted waterways.

Artificial intelligence-based learning model to improve the talents of higher education students towards the digitalization era

Artificial intelligence (AI) technology is a hallmark of the 4.0 revolution. The two main issues in Indonesia are infrastructure that needs to be equipped with technology and intelligence-based curriculum integrated with business and industry programs, and lecturers as educators who do not want to use and develop AI technology in applying guided learning models. This research aims to create a learning model based on AI that will help college students develop their talents while maintaining the Pancasila principles in the age of digitization. This study contains four stages: data collection, data analysis, research analysis outcomes, and validation of research analysis results. This research developed an AI-based learning model for use in higher education consisting of four dimensions: input, process, output, and outcome. The input dimension includes components such as students, lecturers, organizations, and infrastructure ready to adopt AI-based learning models. The process dimension consists of the elements that influence the operation of the AI-based learning model system and the functionality provided by the learning model. The output dimension includes characteristics that may be directly measured and process feedback. Finally, the outcome comprises the predicted outputs from the AI-based learning model.

Training Artificial Neural Networks to Detect Multiple Sclerosis Lesions Using Granulometric Data from Preprocessed Magnetic Resonance Images with Morphological Transformations

The symptoms of multiple sclerosis (MS) are determined by the location of demyelinating lesions in the white matter of the brain and spinal cord. Currently, magnetic resonance imaging (MRI) is the most common tool used for diagnosing MS, understanding the course of the disease, and analyzing the effects of treatments. However, undesirable components may appear during the generation of MRI scans, such as noise or intensity variations. Mathematical morphology (MM) is a powerful image analysis technique that helps to filter the image and extract relevant structures. Granulometry is an image measurement tool for measuring MM that determines the size distribution of objects in an image without explicitly segmenting each object. While several methods have been proposed for the automatic segmentation of MS lesions in MRI scans, in some cases, only simple data preprocessing, such as image resizing to standardize the input dimensions, has been performed before the algorithm training. Therefore, this paper proposes an MRI preprocessing algorithm capable of performing elementary morphological transformations in brain images of MS patients and healthy individuals in order to delete undesirable components and extract the relevant structures such as MS lesions. Also, the algorithm computes the granulometry in MRI scans to describe the size qualities of lesions. Using this algorithm, we trained two artificial neural networks (ANNs) to predict MS diagnoses. By computing the differences in granulometry measurements between an image with MS lesions and a reference image (without lesions), we determined the size characterization of the lesions. Then, the ANNs were evaluated with the validation set, and the performance results (test accuracy = 0.9753; cross-entropy loss = 0.0247) show that the proposed algorithm can support specialists in making decisions to diagnose MS and estimating the disease progress based on granulometry values.

High-level feature-guided attention optimized neural network for neonatal lateral ventricular dilatation prediction.

Periventricular-intraventricular hemorrhage can lead to posthemorrhagic ventricular dilatation or even posthemorrhagic hydrocephalus if not detected promptly. Sequential cranial ultrasound scans are typically used for their diagnoses. Nonetheless, manual image audit has numerous disadvantages. This study aimed to develop a predictive model utilizing modified inception (MI) and high-level feature-guided attention (HFA) modules for predicting neonatal lateral ventricular dilation via ultrasound images. The MI modules reduced input data sizes and dimensions, while the HFA modules effectively delved into semantic information through supervision from high-level feature images to low-level feature images. The process facilitated the accurate identification of dilated lateral ventricles. A total of 710 neonates, corresponding to 1420 lateral ventricles, were recruited in this study. Each lateral ventricle was captured in two images, one on the parasagittal plane and the other on the coronal plane. The combination of anterior horn width, ventricular index, thalamo-occipital distance, and ventricular height served as the gold standard. A lateral ventricle would be considered dilatated if any of these four indices exceeded its upper reference value. These lateral ventricles were randomly split into training and testing sets at a 7:3 ratio. We evaluated the validity of our proposed approach and its competitors across the coronal plane, parasagittal plane, and overall performance. We also determined the impact of subjects' baseline characteristics on the overall performance of the proposed approach. Additionally, ablation analyses were conducted to ensure the efficacy of the proposed approach. Our proposed approach achieved the largest Youden index (0.65, 95% CI: 0.58-0.72), DOR (27.11, 95% CI: 15.89-46.26), area under curves (AUC) of receiver operating characteristic curve (ROC) (0.84, 95% CI: 0.80-0.88), and AUC of precision-recall curve (PRC) (0.81, 95% CI: 0.74-0.86) in the overall performance assessment and ablation analyses. Moreover, it boasted the biggest Cramer's V values on the coronal (Cramer's V=0.488, p<0.001) and parasagittal (Cramer's V=0.713, p<0.001) planes individually. Factors such as left side, male sex, singleton birth, and vaginal delivery were positively correlated with higher performance regarding the proposed algorithm, except for the gestational age. This work provides a novel attention optimized algorithm for rapid and accurate ventricular dilatation predictions. It surpasses the traditional algorithms in terms of validity whether concerning the coronal plane, parasagittal plane, or overall performance. The overall performance of algorithms will be influenced by the baseline characteristics of populations.

Research on a Bearing Fault Diagnosis Method Based on an Improved Wasserstein Generative Adversarial Network

In this paper, an advanced Wasserstein generative adversarial network (WGAN)-based bearing fault diagnosis approach is proposed to bolster the diagnostic efficacy of conventional WGANs and tackle the challenge of selecting optimal hyperparameters while reducing the reliance on sample labeling. Raw vibration signals undergo continuous wavelet transform (CWT) processing to generate time–frequency images that align with the model’s input dimensions. Subsequently, these images are incorporated into a region-based fully convolutional network (R-FCN), substituting the traditional discriminator for feature capturing. The WGAN model is refined through the utilization of the Bayesian optimization algorithm (BOA) to optimize the generator and discriminator’s semi-supervised learning loss function. This approach is verified using the Case Western Reserve University (CWRU) dataset and a centrifugal pump failure experimental dataset. The results showed improvements in data input generalization and fault feature extraction capabilities. By avoiding the need to label large quantities of sample data, the diagnostic accuracy was improved to 98.9% and 97.4%.

Improving the Regional Science and Technology Innovation Ability under the Background of Data Element Marketization: Insights from the Yunnan Province

Under the background of the marketization of data elements, enhancing regional scientific and technological innovation capabilities has become the main driving force for urban sustainable development. It contributes to the sustainable development of the region. Based on the data of the Yunnan statistical yearbook, we built a regional science and technology innovation index system from three dimensions of science and technology innovation input, output, and benefit. Firstly, the entropy method is used to obtain the weight of each index. Secondly, TOPSIS is used to evaluate the 36 index data of each prefecture. Finally, the scores and rankings of the scientific and technological innovation ability of each prefecture are obtained. The research can effectively find the advantages and disadvantages of scientific and technological innovation in the Yunnan Province, provide data support and decision support for regional economic development, and play an important role in promoting the level of technological innovation in modern industries. At the same time, research projects can help various regions allocate various technological innovation resources reasonably, enhance their original innovation capabilities, lead the development of strategic emerging industries and future industries, and accelerate the formation of new quality productivity.

Repmono: a lightweight self-supervised monocular depth estimation architecture for high-speed inference

Self-supervised monocular depth estimation has always attracted attention because it does not require ground truth data. Designing a lightweight architecture capable of fast inference is crucial for deployment on mobile devices. The current network effectively integrates Convolutional Neural Networks (CNN) with Transformers, achieving significant improvements in accuracy. However, this advantage comes at the cost of an increase in model size and a significant reduction in inference speed. In this study, we propose a network named Repmono, which includes LCKT module with a large convolutional kernel and RepTM module based on the structural reparameterisation technique. With the combination of these two modules, our network achieves both local and global feature extraction with a smaller number of parameters and significantly enhances inference speed. Our network, with 2.31MB parameters, shows significant accuracy improvements over Monodepth2 in experiments on the KITTI dataset. With uniform input dimensions, our network’s inference speed is 53.7% faster than R-MSFM6, 60.1% faster than Monodepth2, and 81.1% faster than MonoVIT-small. Our code is available at https://github.com/txc320382/Repmono.

Reinforcement-Learning-Based Multi-UAV Cooperative Search for Moving Targets in 3D Scenarios

Most existing multi-UAV collaborative search methods only consider scenarios of two-dimensional path planning or static target search. To be close to the practical scenario, this paper proposes a path planning method based on an action-mask-based multi-agent proximal policy optimization (AM-MAPPO) algorithm for multiple UAVs searching for moving targets in three-dimensional (3D) environments. In particular, a multi-UAV high–low altitude collaborative search architecture is introduced that not only takes into account the extensive detection range of high-altitude UAVs but also leverages the benefit of the superior detection quality of low-altitude UAVs. The optimization objective of the search task is to minimize the uncertainty of the search area while maximizing the number of captured moving targets. The path planning problem for moving target search in a 3D environment is formulated and addressed using the AM-MAPPO algorithm. The proposed method incorporates a state representation mechanism based on field-of-view encoding to handle dynamic changes in neural network input dimensions and develops a rule-based target capture mechanism and an action-mask-based collision avoidance mechanism to enhance the AM-MAPPO algorithm’s convergence speed. Experimental results demonstrate that the proposed algorithm significantly reduces regional uncertainty and increases the number of captured moving targets compared to other deep reinforcement learning methods. Ablation studies further indicate that the proposed action mask mechanism, target capture mechanism, and collision avoidance mechanism of the AM-MAPPO algorithm can improve the algorithm’s effectiveness, target capture capability, and UAVs’ safety, respectively.