Iterative learning for efficient additive mass production

An interpretability improvement for fuzzy rule bases obtained by the iterative rule learning approach

Single-image dehazing of high-voltage power transmission lines (HPTLs) using deep learning methods confronts two critical challenges: the non-homogeneous haze distribution in HPTL images and the unavailability of paired clear images for supervised training. To overcome these issues, this paper proposes a novel dehaze neural network, named FIF-RSCT-Net, that employs a hybrid supervised-to-unsupervised iterative learning approach according to the characteristic of HPTL single images. The FIF-RSCT-Net incorporates the Spatial–Channel Feature Intersection modules and Residual Separable Convolution Transformers to enhance the feature representation capability. Crucially, this novel architecture could learn more generalized dehazing knowledge that can be transferred from the original image domain to HPTL scenarios. In the dehazing knowledge transformation, an unsupervised iterative learning mechanism based on the Line Segment Detector is designed to optimize the restoration of power transmission lines. The effectiveness of FIF-RSCT-Net on the original image domain is demonstrated in the comparative experiments of the I-Haze, O-Haze, NH-Haze, and SOTS datasets. Our methodology achieves the best average PSNR of 24.647 dB and SSIM of 0.8512. And the qualitative evaluation of unsupervised iterative learning results shows that the missed line segments are exhibited during progressive training iterations.

We present an iterative and unsupervised learning approach for the speaker verification task. In conventional speaker verification, Probabilistic Linear Discriminant Analysis (PLDA) has been widely used as a supervised backend. However, PLDA requires fully labeled training data, which is often difficult to obtain in reality. To automatically retrieve the speaker labels of unlabeled training data, we propose to use the Affinity Propagation (AP) - a clustering method that takes pairwise data similarity as input - to generate the labels for the PLDA modeling. We further propose an iterative refinement strategy that incrementally updates the similarity input of the AP clustering with the previous iteration's PLDA scoring outputs. Moreover, we evaluate the performance of different PLDA scoring methods for the multiple enrollment task and show that the generalized hypothesis testing achieves the best results. Experiments were conducted on the NIST SRE 2010 and the 2014 i-vector challenge database. The results show that our proposed iterative and unsupervised PLDA model learning approach outperformed the cosine similarity baseline by 35% relatively.

This article investigates the consensus tracking problem of the heterogeneous multivehicle systems (MVSs) under a repeatable control environment. First, a unified iterative learning control (ILC) algorithm is presented for all autonomous vehicles, each of which is governed by both discrete- and continuous-time nonlinear dynamics. Then, several consensus criteria for MVSs with switching topology and external disturbances are established based on our proposed distributed ILC protocols. For discrete-time systems, all vehicles can perfectly track to the common reference trajectory over a specified finite time interval, and the corresponding digraphs may not have spanning trees. Existing approaches dealing with the continuous-time systems generally require that all vehicles have strictly identical initial conditions, being too ideal in practice. We relax this unpractical assumption and propose an extra distributed initial state learning protocol such that vehicles can take different initial states, leading to the fact that the finite time tracking is achieved ultimately regardless of the initial errors. Finally, a numerical example demonstrates the effectiveness of our theoretical results.

Encoding desired motions into dynamic movement primitives (DMPs) is a common way for generating compact task representations that are able to handle sensor-based goal adaptations. At the same time, a robot should not only express adaptive motion capabilities at planning level, but use also contact wrench feedback in the adaptation and learning process of the DMP. Despite first approaches exist in this direction, no fully integrated approach has been proposed so far. In this paper, we introduce a new class of admittance-coupled DMPs that addresses environmental changes by including contact wrench feedback dynamics into the DMP formalism. Moreover, a novel iterative learning approach is devised that is based on monitoring the overall system passivity analysis in terms of reference power tracking. Simulations and experimental results with the Kuka LWR robot maintaining a non-rigid contact with the environment (wiping a surface) are shown for supporting the validity of our approach.

In this paper, we propose an error-tracking iterative learning control scheme to tackle the position tracking problem for robot manipulators with random initial errors and iteration-varying reference trajectories. Different from general usual ones, the control strategy in our work is to drive system errors perfectly track the desired error trajectories over the whole time interval as the iteration number increases, by which, the position trajectory and velocity trajectory can respectively track their reference trajectories during the predefined part operation interval. For fulfilling the control design, a new construction method of desired error trajectories is presented to remove the perfect initial resetting condition, which must be satisfied in most traditional iterative learning control algorithms. The uncertainties and disturbances in the robotic system dynamics are compensated by the robust approach and iterative learning approach, combinedly.

Video person re-identification (video Re-ID) plays an important role in surveillance video analysis and has gained increasing attention recently. However, existing supervised methods require vast labeled identities across cameras, resulting in poor scalability in practical applications. Although some unsupervised approaches have been exploited for video Re-ID, they are still in their infancy due to the complex nature of learning discriminative features on unlabelled data. In this paper, we focus on one-shot video Re-ID and present an iterative local-global collaboration learning approach to learning robust and discriminative person representations. Specifically, it jointly considers the global video information and local frame sequence information to better capture the diverse appearance of the person for feature learning and pseudo-label estimation. Moreover, as the cross-entropy loss may induce the model to focus on identity-irrelevant factors, we introduce the variational information bottleneck as a regularization term to train the model together. It can help filter undesirable information and characterize subtle differences among persons. Since accuracy cannot always be guaranteed for pseudo-labels, we adopt a dynamic selection strategy to select part of pseudo-labeled data with higher confidence to update the training set and re-train the learning model. During training, our method iteratively executes the feature learning, pseudo-label estimation, and dynamic sample selection until all the unlabeled data have been seen. Extensive experiments on two public datasets, i.e., DukeMTMC-VideoReID and MARS, have verified the superiority of our model to several cutting-edge competitors.

Iterative learning approach for consensus tracking of partial difference multi-agent systems with control delay under switching topology

Entity Alignment is a research topic that has drawn a lot of attention from practitioners and researchers from many different areas in the last years. Also referred to as Entity Resolution or Instance Matching, it involves aligning different entity representations that refer to the same real-world entity. However, it is still a challenging solution and existing works are based on iterative methods, blocking schemes and learning approaches. Among these latter, recent works have been using embedding models, focusing only on structural information. In this paper, we present an approach for entity alignment that enrich entity embeddings with different literal information. To be able to express complex relationships between properties and involving multiple entities we employ FrameBase Schema, which is based on linguistic frames to create complex mappings from external knowledge bases to a unique schema. Preliminary experiments showed competitive results to the performance of entity alignment task.

Fine grain policy settings in social network sites is becoming a very important requirement for managing user's privacy. Incorrect privacy policy settings can easily lead to leaks in private and personal information. At the same time, being too restrictive would reduce the benefits of online social networks. This is further complicated with the growing adoption of social networks and with the rapid growth in information uploading and sharing. The problem of facilitating policy settings has attracted numerous access control, and human computer interaction researchers. The solutions proposed range from usable interfaces for policy settings to automated policy settings. We propose a fine grained policy recommendation system that is based on an iterative semi-supervised learning approach that uses the social graph propagation properties. Active learning and social graph properties were used to detect the most informative instances to be labeled as training sets. We implemented and tested our approach using real Facebook dataset. We compared our proposed approach to supervised learning and random walk approaches. Our proposed approaches provided high accuracy and precision when compared to the other approaches.

Combining boosting and evolutionary algorithms for learning of fuzzy classification rules

The process of monitoring the events occurring in a computer system or network and analyzing them for sign of intrusions is known as intrusion detection system (IDS). The objective of this paper is to extract fuzzy classification rules for intrusion detection in computer networks. The proposed method is based on the iterative rule learning approach (IRL) to fuzzy rule base system design. The fuzzy rule base is generated in an incremental fashion, in that the evolutionary algorithm optimizes one fuzzy classifier rule at a time. The performance of final fuzzy classification system has been investigated using intrusion detection problem as a high-dimensional classification problem. Results show that the presented algorithm produces fuzzy rules, which can be used to construct a reliable intrusion detection system.

Wireless communication in vehicular networks is highly vulnerable to electromagnetic interference, which seriously undermines the communication effectiveness. Thus, interference estimation and elimination are essential in vehicular communications. To solve the multitasking problem of interference estimation and cancelation within the transformation domain of vehicular communications where they are divided into the consistency and inconsistency based on the different conditions of spectrum sensing, the present study proposes an iterative Bayesian sparse learning approach for simultaneous interference recovery and elimination which applies multiple Bayesian learning for measured signals, and the structured sparse learning method for simultaneous interference mitigation with signal recovery. Furthermore, the updating optimization based on Bayesian estimation framework is performed to facilitate the learning process and improve the performance of interference estimation and signal reconstruction. Simulation results show that the proposed algorithms achieved more rapid convergence and more accurate recovery (31.84% and 27.88% higher in spectrum sensing with consistency and inconsistency, respectively) than conventional algorithms. During all iterations, the proposed algorithms, because of updating optimization, were more robust than conventional ones, but the computational complexity remained largely the same. To explore the correlation between the parameters and the interference recovery performance, we performed experiments, which revealed linear, logarithmic, and exponential regression correlations for particular forms of interference. We also compared the demodulation performance of the proposed algorithm with existing methods, and it turned out the proposed algorithm also obtained higher gains on average in the bit error rate.

This article studies collaborative human- robot output tracking when the desired output is only known to the human but not to the robot controller. The main contribution of this article is to propose and establish convergence conditions for an iterative learning algorithm that updates the robot input using (i) the effect of the human action on the combined human-robot output tracking (which includes the effect of the human-response dynamics) and (ii) data-inferred human-robot models. This allows the iterative learning control (ILC) to be personalized for each individual human operator. Additionally, experimental results are presented to illustrate the iterative learning approach. Results show that, with the proposed approach, the robot can learn to collaboratively track the output with 10.0% error, which is close to twice the robot noise of 4.6% of the desired output. Furthermore, the data-inferred models provided evidence of the effect of the human operator’s dynamics on the co-tracking task.

CNC machines are still suffering from machine blindness. They cannot automatically assess the performance of applied machined tasks. In this article, an approach is made to improve the performance of CNC machining by utilizing on-line vision-based monitoring and control system. To facilitate the integration of computer vision with CNC machines a system is proposed and developed to tackle a number of pinpointed issues that obstruct such integration. A practical executable methodology of these steps is developed to enable their beneficial implementation on lab-scale CNC milling machines. Two different models of bench type CNC machines are employed to generalize the findings. Two cameras are mounted on the machine spindle of each of the two employed CNC machines to provide valid image data according to the cutting direction. Proper selection and activation of relative camera is achieved automatically by the developed system which analyze the most recent conducted tool path movement to decide on which camera is to be activated. In order to assess the machining surface quality and cutting tool status, image data are processed to evaluate resulting tool imprints on the machined surface. An indicating parameter to assess resulting tool imprints is proposed and used. The overall results show the validity of the approach and encourage further development to realize real industrial scale intelligent vision-controlled CNC machines.