Brain-inspired SLAM imitates the neural processing mechanism of the navigation system in the rodent brain. The low computational requirements and minimal storage demands make brain-inspired SLAM particularly well-suited for large-scale environment mapping. However, the increasing size of the mapping environment raises the likelihood of failures in map construction. As far as we know, there is currently no research on collaborative brain-inspired SLAM systems to improve mapping efficiency and error tolerance. In this paper, we propose a real-time collaborative mapping system to fuse local experience maps into the global map. In our method, an effective overlapping region detection based on sequence matching is designed to realize data association between local maps. Through these detected overlapped experience nodes, the relative poses are then obtained. Finally, a continuous map fusion graph relaxation (CMF-GR) algorithm is proposed to merge local maps and correct their drifts. In addition, we present a leader-follower collaborative mapping system to verify the proposed method. We validate the effectiveness and accuracy of our approach on publicly available datasets and real-world environments through comparison experiences and quantitative analysis.

In non-stationary data streams, the challenges of concept drift are further compounded by the issue of Intermediate Verification Latency (IVL), which can impede timely model adaptation. IVL refers to the finite delay between the arrival of data features and their corresponding labels. This delay could pose a significant challenge in adapting models to new concepts, ultimately hindering predictive performance. However, existing IVL approaches exhibit certain limitations. Some approaches passively wait for delayed labels, thereby overlooking temporarily unlabeled data. Other approaches employ pseudo-labeling for immediate model updates, but may risk losing valuable information when reverting model states to rectify previous pseudo-labeling mistakes. To overcome these limitations, we propose a novel approach called Micro-cluster based Immediate Pseudo-Labeling with Oriented Synthetic Correction (MIPLOSC). MIPLOSC leverages micro-cluster systems to effectively capture data distributions, thus facilitating its two core components: immediate pseudo-labeling and oriented synthetic correction. The immediate pseudo-labeling mechanism facilitates immediate utilization of temporarily unlabeled data, and the oriented synthetic correction mechanism enables finergrained rectification from previous erroneous pseudo-labels and concept drift, minimizing the loss of learned information. Experimental studies validated the effectiveness of MIPLOSC in addressing IVL, demonstrating its superiority over competing methods in both space consumption and predictive performance across varying degrees of label delay.

Sub-manifold learning has been widely used to project high-dimensional data into a low-dimensional manifold, preserving the structure of the original data as much as possible. Existing sub-manifold learning methods either learn an embedding manifold that may have different geometric properties as the original data space, or learn a sub-manifold without considering the nonlinear structure of the data. In this paper, we learn a sub-manifold of the original data based on learned prototypes which represent prior knowledge about the intrinsic features of the data. This allows to incorporate the prior knowledge existing in the prototypes to find the suitable sub-manifold. The sub-manifold and the prototypes are jointly learned in a unified cost function via the gradient descent algorithm. The learned prototypes are obtained in the original high-dimensional space and subsequently used to learn a projection matrix to map the high-dimensional data into a lower-dimensional subspace with better separability. The prototypes are relearned in the projected subspace. The relearned low-dimensional prototypes are then working as prior knowledge to induce the learning of a better projection matrix, leading to a better subspace. The proposed subspace learning is realized for data points living in the Riemannian space of symmetric positive definite (SPD) matrices via the generalized learning Riemannian space quantization (GLRSQ) method. Experiments on both synthetic and real-world data sets show the effectiveness of the proposed dimension reduction scheme.

In many classification scenarios, the data to be analyzed can be naturally represented as points living on the curved Riemannian manifold of symmetric positive-definite (SPD) matrices. Due to its non-Euclidean geometry, usual Euclidean learning algorithms may deliver poor performance on such data. We propose a principled reformulation of the successful Euclidean generalized learning vector quantization (GLVQ) methodology to deal with such data, accounting for the nonlinear Riemannian geometry of the manifold through log-Euclidean metric (LEM). We first generalize GLVQ to the manifold of SPD matrices by exploiting the LEM-induced geodesic distance (GLVQ-LEM). We then extend GLVQ-LEM with metric learning. In particular, we study both 1) a more straightforward implementation of the metric learning idea by adapting metric in the space of vectorized log-transformed SPD matrices and 2) the full formulation of metric learning without matrix vectorization, thus preserving the second-order tensor structure. To obtain the distance metric in the full LEM learning (LEML) approaches, two algorithms are proposed. One method is to restrict the distance metric to be full rank, treating the distance metric tensor as an SPD matrix, and readily use the LEM framework (GLVQ-LEML-LEM). The other method is to cast no such restriction, treating the distance metric tensor as a fixed rank positive semidefinite matrix living on a quotient manifold with total space equipped with flat geometry (GLVQ-LEML-FM). Experiments on multiple datasets of different natures demonstrate the good performance of the proposed methods.

Previous class imbalance learning methods are mostly grounded on the assumption that all training data have been labeled, however, is impractical in many real-world applications. The limited amount of labeled instances may produce a classifier with poor generalization. To address the issue, a transfer weighted extreme learning machine (TWELM) classifier is proposed, with the purpose of extracting knowledge from other domains to improve the classification performance of a classifier in a limited labeled target domain. To be specific, a well-tuned weighted extreme learning machine classifier is first learned from source data that has been completely labeled. Subsequently, another extreme learning machine classifier is obtained from the limited labeled target domain data to preserve the target domain structural knowledge and the decision boundary information. Finally, the target classifier is optimized by minimizing the outputs of the two classifiers on unlabeled target data. Experimental results on real-world data sets show that TWELM outperforms existing algorithms on classification accuracy and computation cost.

Many existing works represent signals by covariance matrices and then develop learning methods on the Riemannian symmetric positive-definite (SPD) manifold to deal with such data. However, they summarize each instance with a single covariance matrix, omitting some potential important information, such as the time evolution of the correlation in signals. In this paper, we represent each instance by a sequence of covariance matrices and develop a novel dynamic generalized learning Riemannian space quantization (DGLRSQ) method to deal with such data representations. The proposed DGLRSQ method incorporates short-term memory mechanism in generalized learning Riemannian space quantization (GLRSQ), which is an extension of Euclidean generalized learning vector quantization to deal with SPD matrix-valued data. The proposed method can capture the temporal evolution of the correlation in signals and thus provides better performance to its the counterpart – GLRSQ, which treats each instance as a signal covariance matrix. Empirical investigations on synthetic data and motor imagery EEG data show the superior performance of the proposed method.

In this paper, we develop a new classification method for manifold-valued data in the framework of probabilistic learning vector quantization. In many classification scenarios, the data can be naturally represented by symmetric positive definite matrices, which are inherently points that live on a curved Riemannian manifold. Due to the non-Euclidean geometry of Riemannian manifolds, traditional Euclidean machine learning algorithms yield poor results on such data. In this paper, we generalize the probabilistic learning vector quantization algorithm for data points living on the manifold of symmetric positive definite matrices equipped with Riemannian natural metric (affine-invariant metric). By exploiting the induced Riemannian distance, we derive the probabilistic learning Riemannian space quantization algorithm, obtaining the learning rule through Riemannian gradient descent. Empirical investigations on synthetic data, image data , and motor imagery electroencephalogram (EEG) data demonstrate the superior performance of the proposed method.

Learning vector quantization (LVQ) is a simple and efficient classification method, enjoying great popularity. However, in many classification scenarios, such as electroencephalogram (EEG) classification, the input features are represented by symmetric positive-definite (SPD) matrices that live in a curved manifold rather than vectors that live in the flat Euclidean space. In this article, we propose a new classification method for data points that live in the curved Riemannian manifolds in the framework of LVQ. The proposed method alters generalized LVQ (GLVQ) with the Euclidean distance to the one operating under the appropriate Riemannian metric. We instantiate the proposed method for the Riemannian manifold of SPD matrices equipped with the Riemannian natural metric. Empirical investigations on synthetic data and real-world motor imagery EEG data demonstrate that the performance of the proposed generalized learning Riemannian space quantization can significantly outperform the Euclidean GLVQ, generalized relevance LVQ (GRLVQ), and generalized matrix LVQ (GMLVQ). The proposed method also shows competitive performance to the state-of-the-art methods on the EEG classification of motor imagery tasks.

Visual place recognition in changing environments is a challenging and critical task for autonomous robot navigation. Deep convolutional neural networks (ConvNets) have recently been used as efficient feature extractors and obtained excellent performance in place recognition. However the success of ConvNets' learning highly relies on the availability of large datasets with millions of labeled images, the collection of which is a tedious and costly burden. Thus we develop an unsupervised learning method (the siamese VisNet) to autonomously learn invariant features in changing environments from unlabeled images. The siamese VisNet has two identical branches of sub-networks. With a Hebbian-type of learning rule incorporating a trace of previous activity patterns, the siamese VisNet learns features with increasing invariance in changing environments from layer to layer. Experiments conducted on multiple datasets demonstrate the robustness of the siamese VisNet against viewpoint changes, appearance changes, and joint viewpoint-appearance changes. In addition, the siamese VisNet, with lower complexity in architecture, outperforms the state-of-the-art place recognition ConvNets such as the CaffeNet and the PlaceNet. The proposed siamese VisNet constitutes a biologically plausible yet efficient method for unsupervised place recognition.

Tracking is a common topic in various areas of robotics research. Motivated by the hunting behavior of predators in nature, we propose a prey-predator model for efficient robot tracking. The head direction and speed of the pursuer is automatically adjusted according to the position and velocity of the prey. Under the situation with perception uncertainty, where the actual location of the prey is not observable, the pursuer predicts the location of the prey according to simple inference, an online adaptive autoregressive model, or an online adaptive echo state network. Simulation results demonstrate that the proposed prey-predator model is able to control the pursuer and to track the prey efficiently, even under perception uncertainty. Simple inference gives better results when the motion of the target is piecewise linear, while echo state network is more suitable when the dynamics of the target are more complex. The proposed prey-predator model thus provides an efficient method tracking targets with various statistical nature of trajectories for applications such as underwater robot tracking, human tracking and team formation.

In this letter, we explore the idea of modeling slack variables in support vector machine (SVM) approaches. The study is motivated by SVM+, which models the slacks through a smooth correcting function that is determined by additional (privileged) information about the training examples not available in the test phase. We take a closer look at the meaning and consequences of smooth modeling of slacks, as opposed to determining them in an unconstrained manner through the SVM optimization program. To better understand this difference we only allow the determination and modeling of slack values on the same information--that is, using the same training input in the original input space. We also explore whether it is possible to improve classification performance by combining (in a convex combination) the original SVM slacks with the modeled ones. We show experimentally that this approach not only leads to improved generalization performance but also yields more compact, lower-complexity models. Finally, we extend this idea to the context of ordinal regression, where a natural order among the classes exists. The experimental results confirm principal findings from the binary case.

Echo State Networks (ESNs) have shown great promise in the applications of non-linear time series processing because of their powerful computational ability and efficient training strategy. However, the nature of randomization in the structure of the reservoir causes it be poorly understood and leaves room for further improvements for specific problems. A deterministically constructed reservoir model, Cycle Reservoir with Jumps (CRJ), shows superior generalization performance to standard ESN. However, the weights that govern the structure of the reservoir (reservoir weights) in CRJ model are obtained through exhaustive grid search which is very computational intensive. In this paper, we propose to learn the reservoir weights together with the linear readout weights using a hybrid optimization strategy. The reservoir weights are trained through nonlinear optimization techniques while the linear readout weights are obtained through linear algorithms. The experimental results demonstrate that the proposed strategy of training the CRJ network tremendously improves the computational efficiency without jeopardizing the generalization performance, sometimes even with better generalization performance.

We present novel, efficient, model based kernels for time series data rooted in the reservoir computation framework. The kernels are implemented by fitting reservoir models sharing the same fixed deterministically constructed state transition part to individual time series. The proposed kernels can naturally handle time series of different length without the need to specify a parametric model class for the time series. Compared with most time series kernels, our kernels are computationally efficient. We show how the model distances used in the kernel can be calculated analytically or efficiently estimated. The experimental results on synthetic and benchmark time series classification tasks confirm the efficiency of the proposed kernel in terms of both generalization accuracy and computational speed. This paper also investigates on-line reservoir kernel construction for extremely long time series.

This paper proposes a novel liver cancer identification method based on PSO-SVM. First, the region of interest (ROI) is determined by Lazy-Snapping, and various texture features are extracted from ROI. Afterwards, F-score algorithm is applied to select relevant features, based on which liver cancer classifier is designed by combining parallel Support Vector Machine (SVM) with Particle Swarm Optimization (PSO) algorithm. PSO is used to automatically choose parameters for SVM, and the advantage is that it makes the choice of parameter more objective and avoids the randomicity and subjectivity in the traditional SVM whose parameters are decided through trial and error. The experiment results on real-world datasets show that the proposed parallel PSO-SVM training algorithm improves the prediction accuracy of liver cancer.