Set Of Labels Research Articles

A Branch-and-Price Algorithm for the Online Scheduling of Valet Drivers

In the present paper, the online valet driving problem (OVDP) is studied. In this problem, customers request a valet driving service through the platform, then the valets arrive on e-bikes at the designated pickup location and drive the vehicle to the destination. The key task is to assign the valets effectively for driving orders to minimize the overall cost. To serve that purpose, we first propose a new online scheduling strategy that divides the planning horizon into several rounds with fixed length of time, and each round consists of pooling time and scheduling time. By including the features of online scheduling and the power level of e-bikes, this OVDP becomes more practical but nevertheless challenging. To solve the OVDP, we formulate it into a set partitioning model and design a branch-and-price (B&P) algorithm. To improve the computation efficiency, a label setting algorithm is incorporated to address the pricing subproblem, which is accelerated via a heuristic pricing method. As an essential part of the algorithm design, an artificial column technique and a greedy-based constructive heuristic are implemented to obtain the initial solution. Based on the numerical analysis of various scaled instances, it is verified that the proposed B&P algorithm is not only effective in optimum seeking, but also shows a high level of efficiency in comparison with the off-the-shelf commercial solvers. Furthermore, we also explore the impact of pooling and scheduling time on the OVDP and discover a bowl-shaped trend of the objective value with respect to the two time lengths.

Multi-view prototype-based disambiguation for partial label learning

In this work, we study the multi-view partial label learning (MVPLL) problem, where each instance is depicted by different view features and associated with a set of candidate labels, among which a true label exists but is inaccessible in the training phase. Most existing PLL methods only consider single-view case, which learn view classifier independently and neglect the view correlations, thus can not be applied to solve MVPLL problem. Due to the non-deep framework, traditional MVPLL approach is weak in the representation ability, so its performance is still to be improved. To solve the MVPLL problem, a deep multi-view prototype-based disambiguation approach is proposed in this paper. Specifically, we innovatively employ the deep neural network for multi-view ambiguously-labeled image classification to enhance the representation ability, which makes use of the information fusion between multiple views. To improve the discriminative ability, we propose multi-view prototype-based label disambiguation algorithm. On theoretical aspect, an estimation error bound for view-risk estimator is established, which is shown to be larger than that for fuse-risk estimator. Experiments demonstrate the superiorities of our proposed method in terms of the prediction accuracy.

Neighborhood Weighted Voting-Based Noise Correction for Crowdsourcing

In crowdsourcing scenarios, we can obtain each instance’s multiple noisy labels set from different crowd workers and then use a ground truth inference algorithm to infer its integrated label. Despite the effectiveness of ground truth inference algorithms, a certain level of noise still remains in the integrated labels. To reduce the impact of noise, many noise correction algorithms have been proposed in recent years. To the best of our knowledge, however, nearly all existing noise correction algorithms only exploit each instance’s own multiple noisy label sets but ignore the multiple noisy label sets of its neighbors. Here neighbors refer to the nearest instances found in the feature space based on the distance metric learning. In this article, we propose neighborhood weighted voting-based noise correction (NWVNC). In NWVNC, we at first take advantage of the multiple noisy label sets of each instance’s neighbors (including itself) to estimate the probability that it belongs to its integrated label. Then, we use the estimated probability to identify and filter noise instances and thus obtain a clean set and a noise set. Finally, we train three heterogeneous classifiers on the clean set and correct the noise instances by the consensus voting of three trained classifiers. The experimental results on 34 simulated and two real-world crowdsourced datasets show that NWVNC significantly outperforms all the other state-of-the-art noise correction algorithms used for comparison.

The historical power of the natural science collection of Dominik Bilimek at the University of Natural Resources and Life Sciences Vienna (BOKU)

The scientific world of the nineteenth century was shaped by far-reaching discoveries, expeditions, travel and collection activities as well as by the development of extensive natural scientific social networking. During the 1840s, Dominik Bilimek (1813–1884) arose as a key personality in European natural sciences, with a significant impact on the biological, geoscientific and even archaeological communities. The Bilimek collection at the University of Natural Resources and Life Sciences Vienna encompasses thousands of specimens and provides a large number of original labels with date, locality and companions in the field. This set of collection labels was analysed qualitatively and quantitatively with respect to its historical value to increase our knowledge about the explorer's life and his social network. The study also focuses on Bilimek's so far unknown social links, which were recovered by the processing of the collection, probably resulting in the use of these data for a social network analysis of a nineteenth-century scientific network.

An active learning framework for the low-frequency Non-Intrusive Load Monitoring problem

With the widespread deployment of smart meters worldwide, quantification of energy used by individual appliances via Non-Intrusive Load Monitoring (NILM), i.e., virtual submetering, is an emerging application to inform energy management within buildings. Low-frequency NILM refers to NILM algorithms designed to perform load disaggregation at sampling rates in the order of seconds and minutes, as per smart meter data availability. Recently, many deep learning solutions for NILM have appeared in the literature, with promising results. However, besides requiring large, labelled datasets, the proposed deep learning models are not flexible and usually under-perform when tested in a new environment, affecting scalability. The dynamic nature of appliance ownership and usage inhibits the performance of the developed supervised NILM models and requires large amounts of training data. Transfer learning approaches are commonly used to overcome this issue, but they often assume availability of good quality labelled data from the new environment. We propose an active learning framework, that is able to learn and update the deep learning NILM model from small amounts of data, for transfer to a new environment. We explore the suitability of different types of acquisition functions, which determines which function inputs are most valuable. Finally, we perform a sensitivity analysis of the hyperparameters on model performance. In addition, we propose a modification to the state-of-the-art BatchBALD acquisition function, to address its high computational complexity. Our proposed framework achieves optimal accuracy-labelling effort trade-off with only 5%–15% of the query pool labelled. The results on the REFIT dataset, demonstrate the potential of the proposed active learning to improve transferability and reduce the cost of labelling. Unlike the common approach of retraining the entire model once a new set of labels is provided, we demonstrate that full re-training is not necessary, since a fine-tuning approach can offer a good trade-off between performance achieved and computational resources needed.

MAX CUT in Weighted Random Intersection Graphs and Discrepancy of Sparse Random Set Systems

Let V be a set of n vertices, {mathcal M} a set of m labels, and let {textbf{R}} be an m times n matrix ofs independent Bernoulli random variables with probability of success p; columns of {textbf{R}} are incidence vectors of label sets assigned to vertices. A random instance G(V, E, {textbf{R}}^T {textbf{R}}) of the weighted random intersection graph model is constructed by drawing an edge with weight equal to the number of common labels (namely [{textbf{R}}^T {textbf{R}}]_{v,u}) between any two vertices u, v for which this weight is strictly larger than 0. In this paper we study the average case analysis of Weighted Max Cut, assuming the input is a weighted random intersection graph, i.e. given G(V, E, {textbf{R}}^T {textbf{R}}) we wish to find a partition of V into two sets so that the total weight of the edges having exactly one endpoint in each set is maximized. In particular, we initially prove that the weight of a maximum cut of G(V, E, {textbf{R}}^T {textbf{R}}) is concentrated around its expected value, and then show that, when the number of labels is much smaller than the number of vertices (in particular, m=n^{alpha }, alpha <1), a random partition of the vertices achieves asymptotically optimal cut weight with high probability. Furthermore, in the case n=m and constant average degree (i.e. p = frac{Theta (1)}{n}), we show that with high probability, a majority type randomized algorithm outputs a cut with weight that is larger than the weight of a random cut by a multiplicative constant strictly larger than 1. Then, we formally prove a connection between the computational problem of finding a (weighted) maximum cut in G(V, E, {textbf{R}}^T {textbf{R}}) and the problem of finding a 2-coloring that achieves minimum discrepancy for a set system Sigma with incidence matrix {textbf{R}} (i.e. minimum imbalance over all sets in Sigma ). We exploit this connection by proposing a (weak) bipartization algorithm for the case m=n, p = frac{Theta (1)}{n} that, when it terminates, its output can be used to find a 2-coloring with minimum discrepancy in a set system with incidence matrix {textbf{R}}. In fact, with high probability, the latter 2-coloring corresponds to a bipartition with maximum cut-weight in G(V, E, {textbf{R}}^T {textbf{R}}). Finally, we prove that our (weak) bipartization algorithm terminates in polynomial time, with high probability, at least when p = frac{c}{n}, c<1.

Graph rewriting and relabeling with PBPO+: A unifying theory for quasitoposes

We extend the powerful Pullback-Pushout (PBPO) approach for graph rewriting with strong matching. Our approach, called PBPO+, allows more control over the embedding of the pattern in the host graph, which is important for a large class of rewrite systems. We argue that PBPO+ can be considered a unifying theory in the general setting of quasitoposes, by demonstrating that PBPO+ can define a strict superset of the rewrite relations definable by PBPO, AGREE and DPO. Additionally, we show that PBPO+ is well suited for rewriting labeled graphs and some classes of attributed graphs, by introducing a lattice structure on the label set and requiring graph morphisms to be order-preserving.

Features in children's counting books that lead dyads to both count and label sets during shared book reading.

This study examined how book features influence talk during shared book reading. We used data from a study in which parent-child dyads (n= 157; child's Mage = 43.99 months; 88 girls, 69 boys; 91.72% of parents self-reported as white) were randomly assigned to read two number books. The focus was comparison talk (i.e., talk in which dyads count a set and also label its total), as this type of talk has been shown to promote children's understanding of cardinality. Replicating previous findings, dyads produced relatively low levels of comparison talk. However, book features influenced the talk. Books containing a greater number of numerical representations (e.g., number word, numeral, and non-symbolic set) and a greater word count elicited more comparison talk.

GUDN: A novel guide network with label reinforcement strategy for extreme multi-label text classification

Extreme multi-label text classification (XMTC) is an emerging and essential task in natural language processing. Its objective is to retrieve the most relevant labels for a text from a large set of labels while balancing time and accuracy. Although large-scale pre-trained models have brought new perspectives to this task, more attention should be given to valuable fine-tuned methods and the significant semantic gap between texts and labels. In this paper, we propose a novel guide network (GUDN) with a label reinforcement strategy based on label semantics to help fine-tune pre-trained models for classification. Experimental results demonstrate that GUDN outperforms state-of-the-art methods on Eurlex-4k and achieves competitive results on other popular datasets. In addition, we find that meaningless tokens can harm the Transformer-based model’s classification accuracy in another experiment. We conclude that GUDN is effective in the presence of solid semantics. Our source code is available at https://t.hk.uy/aFSH.

A Branch and Price Algorithm for the Drop-and-Pickup Container Drayage Problem with Empty Container Constraints

This paper addresses the drop-and-pickup container drayage problem with empty container constraints. In this problem, a truck is allowed to drop off the container at the customer and then leave. After the container has been packed/unpacked, the truck returns to pick it up. The problem is further complicated by the fact that empty containers at the depot are often limited in number. This container drayage problem is of great practical importance but seldom investigated. In this paper, we first formulate the problem as a directed graph and then mathematically model it as a mixed-integer linear program (MILP) with the objective of minimizing total travel costs. To solve the MILP effectively, we devise a branch and price algorithm that incorporates several performance enhancement strategies, including three versions of the bi-directional label setting algorithm, preprocessing of time windows and a heuristic for high-quality upper bounds. The experimental results indicate that (1) the proposed algorithm significantly outperforms CPLEX in terms of efficiency and effectiveness, (2) an average cost saving of 9.95∼12.25% can be achieved from the drop-and-pickup mode and (3) the benefit of drop-and-pickup mode increases when the customer density and the fixed cost increase.

A dynamic dispatching problem for autonomous mine trucks in open-pit mines considering endogenous congestion

The introduction of autonomous mine trucks can improve the efficiency, productivity, and safety of open-pit mines, but may require more reasonable dynamic truck dispatching system to guide routes and schedules than manual driven mine trucks. In practical operations, for security consideration, only one autonomous mine truck is allowed to pass through each intersection at a time, which may lead to endogenous congestion at intersections. In this paper, we first introduce the autonomous mine trucks and endogenous congestion at intersections in the dynamic truck dispatching problem. We propose the concept of road resource which denotes the time-extended intersection and give the time–space network with road resources. The rolling horizon framework is applied to model the dynamic feature. In each horizon, the dispatching of autonomous mine trucks is formulated as a mixed integer linear programming model on the time–space network with road resources. To solve the dispatching problem in each horizon, we propose a Lagrangian relaxation algorithm to decompose the problem into a series of shortest path problem, which is solved by the designed label setting algorithm. Moreover, to yield high-quality feasible solution in each iteration of Lagrangian relaxation algorithm, a repairing algorithm with an adaptive initialization strategy is designed. A case study of an open-pit mine in China is involved to show the efficiency of the designed model with road resources and the proposed algorithm. The discussions about the productions, waiting time, travel distance, and serving time are given by sensitivity analysis.

A machine learning-based detection framework against intermittent electricity theft attack

The widespread installation of advanced metering infrastructure (AMI) brings convenience to applications including but not limited to load management and demand response. However, AMI is also at risk of electricity theft and non-technical loss. Using the smart meter data provided by AMI to dig out user electricity consumption behavior is an effective way to construct electricity theft detectors. In this paper, a new intermittent electricity theft attack behavior is presented which switches between committing electricity theft and honestly consuming electricity alternately to skillfully evade the existing detectors. Based on the assumption that the labels of intermittent adversaries are unavailable, a new machine learning-based detection framework is proposed to detect this attack. Initially electricity features are constructed based on time intervals divided by a numerical iteration method. Then light gradient boosting method (LightGBM) is used to classify the normal users and adversaries. Further, the disperse degree of users is designed for capturing the differences between the intermittent adversaries and others. A new 2D label set is then constructed by combining the predicted labels of LightGBM and the disperse degree. Finally, a variational Bayesian Gaussian mixture model is employed based on the 2D label set to sort the users into the normal users and adversaries visually. Results of the case studies show that the presented attack can evade state-of-the-art detectors but still gain high profits. In addition, the proposed machine learning-based detector outperforms state-of-the-art detectors on both persistent attacks and intermittent attacks.

Query-efficient model extraction for text classification model in a hard label setting

Designing a query-efficient model extraction strategy to steal models from cloud-based platforms with black-box constraints remains a challenge, especially for language models. In a more realistic setting, a lack of information about the target model’s internal parameters, gradients, training data, or even confidence scores prevents attackers from easily copying the target model. Selecting informative and useful examples to train a substitute model is critical to query-efficient model stealing. We propose a novel model extraction framework that fine-tunes a pretrained model based on bidirectional encoder representations from transformers (BERT) while improving query efficiency by utilizing an active learning selection strategy. The active learning strategy, incorporating semantic-based diversity sampling and class-balanced uncertainty sampling, builds an informative subset from the public unannotated dataset as the input for fine-tuning. We apply our method to extract deep classifiers with identical and mismatched architectures as the substitute model under tight and moderate query budgets. Furthermore, we evaluate the transferability of adversarial examples constructed with the help of the models extracted by our method. The results show that our method achieves higher accuracy with fewer queries than existing baselines and the resulting models exhibit a high transferability success rate of adversarial examples.

Graph Node Based Interpretability Guided Sample Selection for Active Learning.

While supervised learning techniques have demonstrated state-of-the-art performance in many medical image analysis tasks, the role of sample selection is important. Selecting the most informative samples contributes to the system attaining optimum performance with minimum labeled samples, which translates to fewer expert interventions and cost. Active Learning (AL) methods for informative sample selection are effective in boosting performance of computer aided diagnosis systems when limited labels are available. Conventional approaches to AL have mostly focused on the single label setting where a sample has only one disease label from the set of possible labels. These approaches do not perform optimally in the multi-label setting where a sample can have multiple disease labels (e.g. in chest X-ray images). In this paper we propose a novel sample selection approach based on graph analysis to identify informative samples in a multi-label setting. For every analyzed sample, each class label is denoted as a separate node of a graph. Building on findings from interpretability of deep learning models, edge interactions in this graph characterize similarity between corresponding interpretability saliency map model encodings. We explore different types of graph aggregation to identify informative samples for active learning. We apply our method to public chest X-ray and medical image datasets, and report improved results over state-of-the-art AL techniques in terms of model performance, learning rates, and robustness.

Prior Knowledge Regularized Self-Representation Model for Partial Multilabel Learning.

Partial multilabel learning (PML) aims to learn from training data, where each instance is associated with a set of candidate labels, among which only a part is correct. The common strategy to deal with such a problem is disambiguation, that is, identifying the ground-truth labels from the given candidate labels. However, the existing PML approaches always focus on leveraging the instance relationship to disambiguate the given noisy label space, while the potentially useful information in label space is not effectively explored. Meanwhile, the existence of noise and outliers in training data also makes the disambiguation operation less reliable, which inevitably decreases the robustness of the learned model. In this article, we propose a prior label knowledge regularized self-representation PML approach, called PAKS, where the self-representation scheme and prior label knowledge are jointly incorporated into a unified framework. Specifically, we introduce a self-representation model with a low-rank constraint, which aims to learn the subspace representations of distinct instances and explore the high-order underlying correlation among different instances. Meanwhile, we incorporate prior label knowledge into the above self-representation model, where the prior label knowledge is regarded as the complement of features to obtain an accurate self-representation matrix. The core of PAKS is to take advantage of the data membership preference, which is derived from the prior label knowledge, to purify the discovered membership of the data and accordingly obtain more representative feature subspace for model induction. Enormous experiments on both synthetic and real-world datasets show that our proposed approach can achieve superior or comparable performance to state-of-the-art approaches.

Learning multi-tasks with inconsistent labels by using auxiliary big task

Multi-task learning is to improve the performance of the model by transferring and exploiting common knowledge among tasks. Existing MTL works mainly focus on the scenario where label sets among multiple tasks (MTs) are usually the same, thus they can be utilized for learning across the tasks. However, the real world has more general scenarios in which each task has only a small number of training samples and their label sets are just partially overlapped or even not. Learning such MTs is more challenging because of less correlation information available among these tasks. For this, we propose a framework to learn these tasks by jointly leveraging both abundant information from a learnt auxiliary big task with sufficiently many classes to cover those of all these tasks and the information shared among those partially-overlapped tasks. In our implementation of using the same neural network architecture of the learnt auxiliary task to learn individual tasks, the key idea is to utilize available label information to adaptively prune the hidden layer neurons of the auxiliary network to construct corresponding network for each task, while accompanying a joint learning across individual tasks. Extensive experimental results demonstrate that our proposed method is significantly competitive compared to state-of-the-art methods.

Online Learning for Wearable EEG-Based Emotion Classification

Giving emotional intelligence to machines can facilitate the early detection and prediction of mental diseases and symptoms. Electroencephalography (EEG)-based emotion recognition is widely applied because it measures electrical correlates directly from the brain rather than indirect measurement of other physiological responses initiated by the brain. Therefore, we used non-invasive and portable EEG sensors to develop a real-time emotion classification pipeline. The pipeline trains different binary classifiers for Valence and Arousal dimensions from an incoming EEG data stream achieving a 23.9% (Arousal) and 25.8% (Valence) higher F1-Score on the state-of-art AMIGOS dataset than previous work. Afterward, the pipeline was applied to the curated dataset from 15 participants using two consumer-grade EEG devices while watching 16 short emotional videos in a controlled environment. Mean F1-Scores of 87% (Arousal) and 82% (Valence) were achieved for an immediate label setting. Additionally, the pipeline proved to be fast enough to achieve predictions in real-time in a live scenario with delayed labels while continuously being updated. The significant discrepancy from the readily available labels on the classification scores leads to future work to include more data. Thereafter, the pipeline is ready to be used for real-time applications of emotion classification.

Label modification and bootstrapping for zero-shot cross-lingual hate speech detection

The goal of hate speech detection is to filter negative online content aiming at certain groups of people. Due to the easy accessibility and multilinguality of social media platforms, it is crucial to protect everyone which requires building hate speech detection systems for a wide range of languages. However, the available labeled hate speech datasets are limited, making it difficult to build systems for many languages. In this paper we focus on cross-lingual transfer learning to support hate speech detection in low-resource languages, while highlighting label issues across application scenarios, such as inconsistent label sets of corpora or differing hate speech definitions, which hinder the application of such methods. We leverage cross-lingual word embeddings to train our neural network systems on the source language and apply them to the target language, which lacks labeled examples, and show that good performance can be achieved. We then incorporate unlabeled target language data for further model improvements by bootstrapping labels using an ensemble of different model architectures. Furthermore, we investigate the issue of label imbalance in hate speech datasets, since the high ratio of non-hate examples compared to hate examples often leads to low model performance. We test simple data undersampling and oversampling techniques and show their effectiveness.

An Improved Multilabel k-Nearest Neighbor Algorithm Based on Value and Weight

Multilabel data share important features, including label imbalance, which has a significant influence on the performance of classifiers. Because of this problem, a widely used multilabel classification algorithm, the multilabel k-nearest neighbor (ML-kNN) algorithm, has poor performance on imbalanced multilabel data. To address this problem, this study proposes an improved ML-kNN algorithm based on value and weight. In this improved algorithm, labels are divided into minority and majority, and different strategies are adopted for different labels. By considering the label of latent information carried by the nearest neighbors, a value calculation method is proposed and used to directly classify majority labels. Additionally, to address the misclassification problem caused by a lack of nearest neighbor information for minority labels, weight calculation is proposed. The proposed weight calculation converts distance information with and without label sets in the nearest neighbors into weights. The experimental results on multilabel datasets from different benchmarks demonstrate the performance of the algorithm, especially for datasets with high imbalance. Different evaluation metrics show that the results are improved by approximately 2–10%. The verified algorithm could be applied to a multilabel classification of various fields involving label imbalance, such as drug molecule identification, building identification, and text categorization.

CNN-BiLSTM-Attention: A multi-label neural classifier for short texts with a small set of labels

We propose a CNN-BiLSTM-Attention classifier to classify online short messages in Chinese posted by users on government web portals, so that a message can be directed to one or more government offices. Our model leverages every bit of information to carry out multi-label classification, to make use of different hierarchical text features and the labels information. In particular, our designed method extracts label meaning, the CNN layer extracts local semantic features of the texts, the BiLSTM layer fuses the contextual features of the texts and the local semantic features, and the attention layer selects the most relevant features for each label. We evaluate our model on two public large corpuses, and our high-quality handcraft e-government multi-label dataset, which is constructed by the text annotation tool doccano and consists of 29920 data points. Experimental results show that our proposed method is effective under common multi-label evaluation metrics, achieving micro-f1 of 77.22%, 84.42%, 87.52%, and marco-f1 of 77.68%, 73.37%, 83.57% on these three datasets respectively, confirming that our classifier is robust. We conduct ablation study to evaluate our label embedding method and attention mechanism. Moreover, case study on our handcraft e-government multi-label dataset verifies that our model integrates all types of semantic information of short messages based on different labels to achieve text classification.