Negative Bag Research Articles

Multi-Instance Learning with One Side Label Noise

Multi-instance Learning (MIL) is a popular learning paradigm arising from many real applications. It assigns a label to a set of instances, which is called a bag, and the bag’s label is determined by the instances within it. A bag is positive if and only if it has at least one positive instance. Since labeling bags is more complicated than labeling each instance, we will often face the mislabeling problem in MIL. Furthermore, it is more common that a negative bag has been mislabeled to a positive one, since one mislabeled instance will lead to the change of the whole bag label. This is an important problem that originated from real applications, e.g., web mining and image classification, but little research has concentrated on it as far as we know. In this article, we focus on this MIL problem with one side label noise that the negative bags are mislabeled as positive ones. To address this challenging problem, we propose, to the best our our knowledge, a novel multi-instance learning method with one side label noise. We design a new double weighting approach under traditional framework to characterize the “faithfulness” of each instance and each bag in learning the classifier. Briefly, on the instance level, we employ a sparse weighting method to select the key instances, and the MIL problem with one size label noise is converted to a mislabeled supervised learning scenario. On the bag level, the weights of bags, together with the selected key instances, will be utilized to identify the real positive bags. In addition, we have solved our proposed model by an alternative iteration method with proved convergence behavior. Empirical studies on various datasets have validated the effectiveness of our method.

Grey-based multiple instance learning with multiple bag-representative

Multiple instance learning is a modification in supervised learning that handles the classification of collection instances, which called bags. Each bag contains a number of instances whose features are extracted. In multiple instance learning, the standard assumption is that a positive bag contains at least one positive instance, whereas a negative bag is only comprised of negative instances. The complexity of multiple instance learning relies heavily on the number of instances in the training datasets. Since we are usually confronted with a large instance space, it is important to design efficient instance selection techniques to speed up the training process, without compromising the performance. Firstly, a multiple instance learning model of support vector machine based on grey relational analysis is proposed in this paper. The data size can be reduced, and the importance of instances in the bag can be preliminarily judged. Secondly, this paper introduces an algorithm with the bag-representative selector that trains the support vector machine based on bag-level information. Finally, this paper shows how to generalize the algorithm for binary multiple instance learning to multiple class tasks. The experimental study evaluates and compares the performance of our method against 8 state-of-the-art multiple instance methods over 10 datasets, and then demonstrates that the proposed approach is competitive with the state-of-art multiple instance learning methods.

An embarrassingly simple approach to neural multiple instance classification

Multiple Instance Learning (MIL) is a weak supervision learning paradigm that allows modeling of machine learning problems in which labels are not available for individual examples but only for groups of examples called bags. A positive bag may contain one or more positive examples but it is not known which examples in the bag are positive. All examples in a negative bag belong to the negative class. Such problems arise frequently in fields of computer vision, medical image processing and bioinformatics. Many neural network-based solutions have been proposed in the literature for MIL. However, almost all of them rely on introducing specialized blocks and connectivity in their architectures. In this paper, we present a simple and effective approach to Multiple Instance Learning in neural networks. We propose a simple bag-level ranking loss function that allows Multiple Instance Classification in any neural architecture. We have demonstrated the effectiveness of our proposed method for popular MIL benchmark datasets. Additionally, we have also tested the performance of our method in convolutional neural networks used to model an MIL problem derived from the well-known MNIST dataset. Results show that despite being simpler, our proposed scheme is comparable or better than existing methods in the literature in practical scenarios. Python code files for all the experiments can be found at https://github.com/amina01/ESMIL

Association of a Multidisciplinary Care Approach With the Quality of Care After Pediatric Tracheostomy

Incidence of tracheostomy placement in children is increasing, and these children continue to have high incidences of morbidity and mortality. A multidisciplinary tracheostomy program may help improve the quality of care received by these patients. To determine whether implementation of a multidisciplinary tracheostomy program can improve the care of children who received a tracheostomy through reduction in tracheostomy-related adverse events (TRAEs), improved tracheostomy education, and caregiver preparedness. A prospective cohort study was conducted from January 2015 to June 2018 at a pediatric tertiary referral center in Boston, Massachusetts. The participants included 700 children who had received a tracheostomy, most of whom were aged birth to 18 years, but some patients with congenital disorders were much older. Institution of a multidisciplinary tracheostomy team (MDT) whose activities included conducting staff meetings, organizing outpatient clinics, conducting inpatient tracheostomy ward rounds, and conducting inpatient tracheostomy rounds at a local rehabilitation hospital. Quality improvement initiatives included monitoring standardized TRAEs and distributing standardized tracheostomy "go-bags." Reduction of TRAEs and improved caregiver preparedness through distribution of tracheostomy go-bags were assessed following the establishment of a multidisciplinary tracheostomy program. In total, 700 children who had received a tracheostomy during the study period were actively followed up by the MDT. Of these children, 378 (54.0%) were males and 322 (46.0%) were females; mean (SD) age was 4.1 (6.1) years. More than 60 new pediatric tracheostomies were performed annually at the referral center. Reported TRAEs were reduced by 43.0% from the first to the third year after the implementation of a standardized, closed-loop monitoring system (from a mean [SD] of 6.1 [5.2] TRAEs per 1000 inpatient tracheostomy-days in 2015 to a mean [SD] of 4.0 [2.5] in 2018). The most common TRAE was unplanned decannulation, which occurred 64 times during the study period. On average, 10 patients were seen in each monthly multidisciplinary tracheostomy clinic. Clinic interventions included continuing care (146 [52.5%]), communication enhancement (67 [23.6%]), plans for decannulation (52 [18.6%]), and referrals for comorbidities (13 [4.6%]). Approximately 19 inpatients were seen during biweekly rounds and 8 during monthly rounds at a local rehabilitation hospital. A total of 297 patients received standardized tracheostomy go-bags, and more than 70 positive bag checks were performed in the monthly MDT clinics. A positive bag check refers to the incidence when a family is given a go-bag and also uses it. In contrast, a negative bag check refers to when a family is given a go-bag but neither brings it to the clinic nor acknowledges that they use it. This study's findings suggest that a multidisciplinary tracheostomy program may be a powerful tool for enhancing patient safety and quality improvement. Ongoing studies will develop measurable pediatric tracheostomy outcome metrics and assess long-term outcomes.

Discriminative Video Representation Learning Using Support Vector Classifiers.

Most popular deep models for action recognition in videos generate independent predictions for short clips, which are then pooled heuristically to assign an action label to the full video segment. As not all frames may characterize the underlying action-indeed, many are common across multiple actions-pooling schemes that impose equal importance on all frames might be unfavorable. In an attempt to tackle this problem, we propose discriminative pooling, based on the notion that among the deep features generated on all short clips, there is at least one that characterizes the action. To identify these useful features, we resort to a negative bag consisting of features that are known to be irrelevant, for example, they are sampled either from datasets that are unrelated to our actions of interest or are CNN features produced via random noise as input. With the features from the video as a positive bag and the irrelevant features as the negative bag, we cast an objective to learn a (nonlinear) hyperplane that separates the unknown useful features from the rest in a multiple instance learning formulation within a support vector machine setup. We use the parameters of this separating hyperplane as a descriptor for the full video segment. Since these parameters are directly related to the support vectors in a max-margin framework, they can be treated as a weighted average pooling of the features from the bags, with zero weights given to non-support vectors. Our pooling scheme is end-to-end trainable within a deep learning framework. We report results from experiments on eight computer vision benchmark datasets spanning a variety of video-related tasks and demonstrate state-of-the-art performance across these tasks.

A Lagrangian Relaxation Approach for Binary Multiple Instance Classification.

In the standard classification problems, the objective is to categorize points into different classes. Multiple instance learning (MIL), instead, is aimed at classifying bags of points, each point being an instance. The main peculiarity of a MIL problem is that, in the learning phase, only the label of each bag is known whereas the labels of the instances are unknown. We discuss an instance-level learning approach for a binary MIL classification problem characterized by two classes of instances, positive and negative, respectively. In such a problem, a negative bag is constituted only by negative instances, while a bag is positive if it contains at least one positive instance. We start from a mixed integer nonlinear optimization model drawn from the literature and the main result we obtain is to prove that a Lagrangian relaxation approach, equipped with a dual ascent scheme, allows us to obtain an optimal solution of the original problem. The relaxed problem is tackled by means of a block coordinate descent (BCD) algorithm. We provide, finally, the results of our implementation on some benchmark data sets.

Comparison of positioning accuracy between personalized polyurethane foam with wing boards and negative pressure vacuum bag in radiotherapy for lung cancer

Objective To compare the positioning accuracy between personalized polyurethane foam with wing boards and negative pressure vacuum bag in radiotherapy for lung cancer using kilovoltage cone beam computed tomography (CBCT). Methods Thirty-nine patients with lung cancer who received chest radiotherapy in our hospital from 2015 to 2016 were enrolled as subjects. In those patients, 20 were immobilized by negative pressure vacuum bags (VB group) and the others by personalized polyurethane foam with wing boards (PF group). CBCT images were acquired weekly and registered with planning CT images. Setup errors in the left-right, superior-inferior, and anterior-posterior directions, three-dimensional (3D) displacement vector, and setup time were recorded. The margin of the planning target volume (PTV) was calculated using the Van Herk formula (2.5∑+ 0.7σ). Between-group comparison was made by paired t test. Results The PF group had a significant smaller setup error in the y-direction than the VB group (2.54±1.79 vs. 3.19±2.14 mm, P=0.03), while there were no significant differences in setup errors in the x-or z-direction between the two groups (1.80±1.48 vs. 1.90±1.41 mm, P=0.46; 2.14±1.75 vs. 2.25±1.75 mm, P=0.35). There were no significant differences in rotational setup errors in the Rx-, Ry-, or Rz-direction between the two groups (1.15°±0.76°vs. 1.15°±0.85°, P=0.50; 0.71°±0.60°vs. 0.72°±0.43°, P=0.45; 0.62°±0.54° vs. 0.46°±0.30°, P=0.06). The PTV margins in the x-, y-, and z-directions were expanded by 5.56, 8.57, and 7.02 mm, respectively, in the PF group, and by 5.62, 9.27, and 7.23 mm, respectively, in the VB group. The proportion of patients with 3D displacement vectors larger than 5 mm was 40% in the PF group and 45% in the VB group. Conclusions For patients undergoing radiotherapy for lung cancer, personalized polyurethane foam with wing boards can, to a certain extent, reduce the setup error in the superior-inferior direction and PTV margin expansion. Key words: Lung neoplasms/radiotherapy; Polyurethane foam; Vacuum bag; Setup errors; Margin

Stochastic learning of multi-instance dictionary for earth mover’s distance-based histogram comparison

Dictionary plays an important role in multi-instance data representation. It maps bags of instances to histograms. Earth mover’s distance (EMD) is the most effective histogram distance metric for the application of multi-instance retrieval. However, up to now, there is no existing multi-instance dictionary learning methods designed for EMD-based histogram comparison. To fill this gap, we develop the first EMD-optimal dictionary learning method using stochastic optimization method. In the stochastic learning framework, we have one triplet of bags, including one basic bag, one positive bag, and one negative bag. These bags are mapped to histograms using a multi-instance dictionary. We argue that the EMD between the basic histogram and the positive histogram should be smaller than that between the basic histogram and the negative histogram. Base on this condition, we design a hinge loss. By minimizing this hinge loss and some regularization terms of the dictionary, we update the dictionary instances. The experiments over multi-instance retrieval applications shows its effectiveness when compared to other dictionary learning methods over the problems of medical image retrieval and natural language relation classification.

Multiple-instance learning based decision neural networks for image retrieval and classification

The revolutionary Internet and digital technologies have spawned a need for technology that can organize abundantly available digital images for easy categorization and retrieval. Hence, content-based image retrieval (CBIR) has become one of the most active research areas for the last few decades. However, it is still an open issue to narrow down the gap between the high level semantics in the human minds and the low level features computable by machines. This paper proposes a multiple-instance learning based decision neural network (MI-BDNN) that attempts to bridge the semantic ga in CBIR. Multiple-instance learning (MIL) is a variation of supervised learning, where the training set is composed of many bags, and each bag contains many instances. If a bag contains at least one positive instance, it is labelled as a positive bag; otherwise, it is labelled as a negative bag. A novel discriminant function and learning schemes are employed in the MI-BDNN to learn the concept from the training bags. The proposed approach considers the image retrieval problem as a MIL problem, where a user׳s preferred image concept is learned by training MI-BDNN with a set of exemplar images, each of which is labelled as conceptual related (positive) or conceptual unrelated (negative) image. The MI-BDNN based CBIR system is developed, and the results of the experiments showed that MI-BDNN can successfully be used for real image retrieval and classification problems.

Multiple kernel-based dictionary learning for weakly supervised classification

In this paper, we develop a multiple instance learning (MIL) algorithm using the dictionary learning framework where the labels are given in the form of positive and negative bags, with each bag containing multiple samples. A positive bag is guaranteed to have only one positive class sample while all the samples in a negative bag belong to the negative class. Given positive and negative bags of data, our method learns appropriate feature space to select positive samples from the positive bags as well as optimal dictionaries to represent data in these bags. We apply this method for digit recognition, action recognition, and gender recognition tasks and demonstrate that the proposed method is robust and can perform significantly better than many competitive two class MIL classification algorithms.

Bag Constrained Structure Pattern Mining for Multi-Graph Classification

This paper formulates a multi-graph learning task. In our problem setting, a bag contains a number of graphs and a class label. A bag is labeled positive if at least one graph in the bag is positive, and negative otherwise. In addition, the genuine label of each graph in a positive bag is unknown, and all graphs in a negative bag are negative. The aim of multi-graph learning is to build a learning model from a number of labeled training bags to predict previously unseen test bags with maximum accuracy. This problem setting is essentially different from existing multi-instance learning (MIL), where instances in MIL share well-defined feature values, but no features are available to represent graphs in a multi-graph bag. To solve the problem, we propose a Multi-Graph Feature based Learning (gMGFL) algorithm that explores and selects a set of discriminative subgraphs as features to transfer each bag into a single instance, with the bag label being propagated to the transferred instance. As a result, the multi-graph bags form a labeled training instance set, so generic learning algorithms, such as decision trees, can be used to derive learning models for multi-graph classification. Experiments and comparisons on real-world multi-graph tasks demonstrate the algorithm performance.

Boosting for multi-graph classification.

In this paper, we formulate a novel graph-based learning problem, multi-graph classification (MGC), which aims to learn a classifier from a set of labeled bags each containing a number of graphs inside the bag. A bag is labeled positive, if at least one graph in the bag is positive, and negative otherwise. Such a multi-graph representation can be used for many real-world applications, such as webpage classification, where a webpage can be regarded as a bag with texts and images inside the webpage being represented as graphs. This problem is a generalization of multi-instance learning (MIL) but with vital differences, mainly because instances in MIL share a common feature space whereas no feature is available to represent graphs in a multi-graph bag. To solve the problem, we propose a boosting based multi-graph classification framework (bMGC). Given a set of labeled multi-graph bags, bMGC employs dynamic weight adjustment at both bag- and graph-levels to select one subgraph in each iteration as a weak classifier. In each iteration, bag and graph weights are adjusted such that an incorrectly classified bag will receive a higher weight because its predicted bag label conflicts to the genuine label, whereas an incorrectly classified graph will receive a lower weight value if the graph is in a positive bag (or a higher weight if the graph is in a negative bag). Accordingly, bMGC is able to differentiate graphs in positive and negative bags to derive effective classifiers to form a boosting model for MGC. Experiments and comparisons on real-world multi-graph learning tasks demonstrate the algorithm performance.

Learning Instance Concepts from Multiple-Instance Data with Bags as Distributions

We analyze and evaluate a generative process for multiple-instance learning (MIL) in which bags are distributions over instances. We show that our generative process contains as special cases generative models explored in prior work, while excluding scenarios known to be hard for MIL. Further, under the mild assumption that every negative instance is observed with nonzero probability in some negative bag, we show that it is possible to learn concepts that accurately label instances from MI data in this setting. Finally, we show that standard supervised approaches can learn concepts with low area-under-ROC error from MI data in this setting. We validate this surprising result with experiments using several synthetic and real-world MI datasets that have been annotated with instance labels.

Online MIL tracking with instance-level semi-supervised learning

In this paper we propose an online multiple instance boosting algorithm with instance-level semi-supervised learning, termed SemiMILBoost, to achieve robust object tracking. Our work revisits the multiple instance learning (MIL) formulation to alleviate the drifting problem in tracking, which addresses two key issues in the existing MIL based tracking-by-detection methods, i.e., the unselective treatment of instances in the positive bag during weak classifier updating and the lack of object prior knowledge in instance modeling. We tackle both issues in a principled way by using a robust SemiMILBoost algorithm, which treats instances in the positive bag as unlabeled while the ones in the negative bag as negative. To improve the discriminability of weak classifiers online, we iteratively update them with the pseudo-labels and importance of all instances in the positive bag, which are predicted by employing the instance-level semi-supervised learning technique with object prior knowledge during boosting. Experimental results demonstrate that our proposed algorithm outperforms the state-of-the-art tracking methods on several challenging video sequences.

Hierarchical Sampling for Multi-Instance Ensemble Learning

In this paper, we propose a Hierarchical Sampling-based Multi-Instance ensemble LEarning (HSMILE) method. Due to the unique multi-instance learning nature, a positive bag contains at least one positive instance whereas samples (instance and sample are interchangeable terms in this paper) in a negative bag are all negative, simply applying bootstrap sampling to individual bags may severely damage a positive bag because a sampled positive bag may not contain any positive sample at all. To solve the problem, we propose to calculate probable positive sample distributions in each positive bag and use the distributions to preserve at least one positive instance in a sampled bag. The hierarchical sampling involves inter- and intrabag sampling to adequately perturb bootstrap sample sets for multi-instance ensemble learning. Theoretical analysis and experiments confirm that HSMILE outperforms existing multi-instance ensemble learning methods.

Study on image retrieval system base on multi-objective and multi-instance learning

In this paper, the multi-instance learning algorithm is improved under the image retrieval framework based on contents, and the improved multi-instance learning algorithm is applied to image retrieval to better handle the ambiguity of the image. In this method, the image is used as the multi-instance bag and is divided into multiple instances by image segmentation algorithm, and then the multi-instance learning is performed with the multi-objective-diverse-density algorithm. The learning results are ordered by image similarity using the vector space model. Finally, relevant feedback is given in accordance with the positive bag and negative bag chosen by the user to provide satisfactory results to the user.

A Study on Quark-Gluon Plasma Equation of State Using Generalized Uncertainty Principle

The effects of Generalized Uncertainty Principle, which has been predicted by various theories of quantum gravity replacing the Heisenberg’s uncertainty principle near the Planck scale, on the thermodynamics of ideal Quark-Gluon Plasma (QGP) consisting of two and three flavors are included. There is a clear effect on thermodynamical quantities like the pressure and the energy density which means that a different effect from quantum gravity may be used in enhancement the theoretical results for Quark-Gluon Plasma state of matter. This effect looks like the technique used in lattice QCD simulation. We determine the value of the bag parameter from fitting lattice QCD data and a physical interpretation to the negative bag pressure is introduced.

Improving Web Image Search by Bag-Based Reranking

Given a textual query in traditional text-based image retrieval (TBIR), relevant images are to be reranked using visual features after the initial text-based search. In this paper, we propose a new bag-based reranking framework for large-scale TBIR. Specifically, we first cluster relevant images using both textual and visual features. By treating each cluster as a "bag" and the images in the bag as "instances," we formulate this problem as a multi-instance (MI) learning problem. MI learning methods such as mi-SVM can be readily incorporated into our bag-based reranking framework. Observing that at least a certain portion of a positive bag is of positive instances while a negative bag might also contain positive instances, we further use a more suitable generalized MI (GMI) setting for this application. To address the ambiguities on the instance labels in the positive and negative bags under this GMI setting, we develop a new method referred to as GMI-SVM to enhance retrieval performance by propagating the labels from the bag level to the instance level. To acquire bag annotations for (G)MI learning, we propose a bag ranking method to rank all the bags according to the defined bag ranking score. The top ranked bags are used as pseudopositive training bags, while pseudonegative training bags can be obtained by randomly sampling a few irrelevant images that are not associated with the textual query. Comprehensive experiments on the challenging real-world data set NUS-WIDE demonstrate our framework with automatic bag annotation can achieve the best performances compared with existing image reranking methods. Our experiments also demonstrate that GMI-SVM can achieve better performances when using the manually labeled training bags obtained from relevance feedback.

Vacuum energy in the bag model

The vacuum energy of the Yang-Mills field is examined for the conditions of the bag model. The dominance of high-frequency effects results in a vacuum energy that decomposes naturally into a volume energy, a surface energy, and higher shape energies. These quantities are identified with the parameters of the bag model. The imposition of confining boundary conditions for all frequencies is shown to be inconsistent since this would result in the bag constant and certain of the shape tensions being infinite. The manner in which the boundary conditions should be relaxed at high frequency is discussed. The most naive procedure for relaxing the boundary conditions, which is to apply confining conditions only on modes of frequency less than some cutoff frequency, results in a negative bag constant and surface tension and would render the vacuum unstable against the spontaneous breaking of Poincaré invariance. Consideration of the manner by which the interacting electromagnetic field avoids a similar instability suggests that a more realistic way to relax the boundary conditions on the bag surface is to endow the vacuum exterior to the bag with a frequency-dependent dielectric constant and magnetic permeability. In this picture the stability of the vacuum is restored, the surface tension is finite and positive, and the bag constant is zero at least to lowest order in the coupling. It is pointed out that the fermion contributions to the bag constant and the surface tension may relate to the spontaneous breaking of chiral invariance. The aim throughout is to examine the bag model, as it relates to vacuum energy, strictly in its own terms with an emphasis on questions of principle. All too often is heard the alibi that since the theory itself is only approximate, the mathematics need be no better. In truth the opposite follows. Granted that the model represents but a part of nature, we are to find what such an ideal picture implies, a result strictly derived serves to test the model; a false result proves nothing but the failure of the theorist. To call an error by a sweeter name does not correct it. The oversimplification or extension afforded by the model is not error: the model, if well made, shows at least how the universe might behave, but logical errors bring us no closer to the reality of any universe. (Truesdell and Toupin 1960).