Training Bags Research Articles

Addressing catastrophic forgetting in payload parameter identification using incremental ensemble learning

Collaborative robots (cobots) are increasingly integrated into Industry 4.0 dynamic manufacturing environments that require frequent system reconfiguration due to changes in cobot paths and payloads. This necessitates fast methods for identifying payload inertial parameters to compensate the cobot controller and ensure precise and safe operation. Our prior work used Incremental Ensemble Model (IEM) to identify payload parameters, eliminating the need for an excitation path and thus removing the separate identification step. However, this approach suffers from catastrophic forgetting. This paper introduces a novel incremental ensemble learning method that addresses the problem of catastrophic forgetting by adding a new weak learner to the ensemble model for each new training bag. Moreover, it proposes a new classification model that assists the ensemble model in identifying which weak learner provides the most accurate estimation for new input data. The proposed method incrementally updates the identification model while the cobot navigates any task path, maintaining accuracy on old weak learner even after updating with new data. Validation performed on the Franka Emika cobot showcases the model’s superior accuracy and adaptability, effectively eliminating the problem of catastrophic forgetting.

DCAMIL: Eye-tracking guided dual-cross-attention multi-instance learning for refining fundus disease detection

Deep neural networks (DNNs) have facilitated the development of computer-aided diagnosis (CAD) systems for fundus diseases, helping ophthalmologists to reduce missed diagnoses and misdiagnosis rates. However, the majority of CAD systems are data-driven, but lack the prior medical knowledge that can be performance-friendly. In this regard, we innovatively proposed a human-in-the-loop (HITL) CAD system by leveraging ophthalmologists’ eye-tracking information. Concretely, the HITL CAD system was implemented on the multi-instance learning (MIL), where clinicians’ gaze maps were beneficial to cherry-pick diagnosis-related instances. Furthermore, the dual-cross-attention MIL (DCAMIL) network was utilized to curb the adverse effects of noisy instances. Meanwhile, both the sequence augmentation (SA) module and the domain adversarial network (DAN) were introduced to enrich and standardize the instances in the training bag, respectively, thereby enhancing the robustness of our method. We conduct comparative experiments on our newly-constructed datasets (namely, AMD-Gaze and DR-Gaze) for the AMD and early DR detection, respectively. Rigorous experiments demonstrate the feasibility of our HITL CAD system and the superiority of the proposed DCAMIL, which fully exploits ophthalmologists’ eye-tracking information. These investigations indicate that clinicians’ gaze maps, as prior medical knowledge, is potential to contribute to the CAD systems of clinical diseases.

Multiple-Instance Learning From Unlabeled Bags With Pairwise Similarity

In <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">multiple-instance learning</i> (MIL), each training example is represented by a bag of instances. A training bag is either negative if it contains no positive instances or positive if it has at least one positive instance. Previous MIL methods generally assume that training bags are fully labeled. However, the exact labels of training examples may not be accessible, due to security, confidentiality, and privacy concerns. Fortunately, it could be easier for us to access the pairwise similarity between two bags (indicating whether two bags share the same label or not) and unlabeled bags, as we do not need to know the underlying label of each bag. In this paper, we provide the first attempt to investigate MIL from only similar-dissimilar-unlabeled bags. To solve this new MIL problem, we first propose a strong baseline method that trains an instance-level classifier by employing an unlabeled-unlabeled learning strategy. Then, we also propose to train a bag-level classifier based on a convex formulation and theoretically derive a generalization error bound for this method. Comprehensive experimental results show that our instance-level classifier works well, while our bag-level classifier even has better performance.

L₁ Sparsity-Regularized Attention Multiple-Instance Network for Hyperspectral Target Detection.

Attention-based deep multiple-instance learning (MIL) has been applied to many machine-learning tasks with imprecise training labels. It is also appealing in hyperspectral target detection, which only requires the label of an area containing some targets, relaxing the effort of labeling the individual pixel in the scene. This article proposes an L1 sparsity-regularized attention multiple-instance neural network (L1-attention MINN) for hyperspectral target detection with imprecise labels that enforces the discrimination of false-positive instances from positively labeled bags. The sparsity constraint applied to the attention estimated for the positive training bags strictly complies with the definition of MIL and maintains better discriminative ability. The proposed algorithm has been evaluated on both simulated and real-field hyperspectral (subpixel) target detection tasks, where advanced performance has been achieved over the state-of-the-art comparisons, showing the effectiveness of the proposed method for target detection from imprecisely labeled hyperspectral data.

Uncertainty Ordinal Multi-Instance Learning for Breast Cancer Diagnosis.

Ordinal multi-instance learning (OMIL) deals with the weak supervision scenario wherein instances in each training bag are not only multi-class but also have rank order relationships between classes, such as breast cancer, which has become one of the most frequent diseases in women. Most of the existing work has generally been to classify the region of interest (mass or microcalcification) on the mammogram as either benign or malignant, while ignoring the normal mammogram classification. Early screening for breast disease is particularly important for further diagnosis. Since early benign lesion areas on a mammogram are very similar to normal tissue, three classifications of mammograms for the improved screening of early benign lesions are necessary. In OMIL, an expert will only label the set of instances (bag), instead of labeling every instance. When labeling efforts are focused on the class of bags, ordinal classes of the instance inside the bag are not labeled. However, recent work on ordinal multi-instance has used the traditional support vector machine to solve the multi-classification problem without utilizing the ordinal information regarding the instances in the bag. In this paper, we propose a method that explicitly models the ordinal class information for bags and instances in bags. Specifically, we specify a key instance from the bag as a positive instance of bags, and design ordinal minimum uncertainty loss to iteratively optimize the selected key instances from the bags. The extensive experimental results clearly prove the effectiveness of the proposed ordinal instance-learning approach, which achieves 52.021% accuracy, 61.471% sensitivity, 47.206% specificity, 57.895% precision, and an 59.629% F1 score on a DDSM dataset.

The reasons for the gap between the results of the test of secondary school students in Taif and the results of the general and achievement test: أسباب الفجوة بين نتائج اختبار طلاب المرحلة الثانوية بالطائف وبين نتائج اختبار القدرات العامة والتحصيلي

The current study aimed at the reasons for the gap between the results of the test of high school students in Taif and the results of the general aptitude and achievement test from the viewpoint of students, teachers, educational supervisors and school leaders To achieve the objectives of the study, the researcher used the descriptive and analytical approach. Two questionnaires were prepared, a questionnaire for students, and the other was distributed to teachers, supervisors and educational leaders. The study sample reached (343) by (193) students and (150) teachers, an educational supervisor and a school leader. The gap regarding students is ranked according to importance as follows: In the first place, fear of the tests leads to failure to perform with an average of (4.1), and in the second place, "students were not trained in the school on tests similar to the aptitude and achievement tests" with an average of (3.97). With regard to teachers, educational supervisors and school leaders, arranged in order of importance as follows: In the first place is "deficiency in the inclusion of samples of aptitude tests and achievement tests in the curriculum" with an average of (4.49), and in the second place is a deficiency in conducting periodic experimental tests for students during the year for training on exams Achievement abilities and tests, with an average of (4.43). In light of the results of the study, the researcher presented a set of recommendations and proposals, the most important of which are: Exercise in awe, attract trainers of the offered mental skills specialists in measurement and evaluation, design tight training bags that specialize in testing general abilities.

Learning With Label Proportions by Incorporating Unmarked Data.

Learning with label proportions (LLP) deals with the problem that the training data are provided as bags, where the label proportions of training bags rather than the labels of individual training instances are accessible. Existing LLP studies assume that the label proportions of all training bags are accessible. However, in many applications, it is time-consuming to mark all training bags with label proportions, which leads to the problem of learning with both marked and unmarked bags, namely, semisupervised LLP (SLLP). In this work, we propose semisupervised proportional support vector machine (SS- ∝ SVM), which extends the proportional SVM ( ∝ SVM) model to its semisupervised version. To the best of our knowledge, SS- ∝ SVM is the first attempt to cope with the SLLP problem. Two realizations, alter-SS- ∝ SVM and conv-SS- ∝ SVM, which are based on alternating optimization and convex relaxation, respectively, are developed to solve the proposed SS- ∝ SVM model. Moreover, we design a cutting plane (CP) method to optimize conv-SS- ∝ SVM with a guaranteed convergence rate and present a fast accelerated proximal gradient method to solve the multiple kernel learning subproblem in conv-SS- ∝ SVM efficiently. Empirical experiments not only justify the superiority of SS- ∝ SVM over its supervised counterpart in classification accuracy but also demonstrate the high competitive computational efficiency of the CP optimization of conv-SS- ∝ SVM.

Witness detection in multi-instance regression and its application for age estimation

The huge data resource on the Web provides us with an emerging chance to solve the lack of training sample problem that lasting for years in facial age estimation. We show that the web images assisted age estimation can be modeled into a Multiple Instance Regression (MIR) problem. Different from the traditional Multiple Instance Learning (MIL) problem that deals with the bag-level classification task, we model the age estimation problem as an instance-level task. To this end, it is essential to reveal the latent instance labels from all bags. In this paper, we propose a novel algorithm named Witness Detecting Multi-instance Regression (WDMR) that can find all possible positive instances from training bags and use them to train an instance-level regressor. Considering the connection between neighbor relationship and inter-/intra- class differences among instances, we develop a Supervised Citer k-Nearest Neighbor (SC-kNN) graph and a sparse voting strategy to address these problems within a joint learning framework. Experimental results on synthetic and real-world datasets have verified the advantages of our method compared with other state-of-the-art MIL approaches.

Software Effort Interval Prediction via Bayesian Inference and Synthetic Bootstrap Resampling

Software effort estimation (SEE) usually suffers from inherent uncertainty arising from predictive model limitations and data noise. Relying on point estimation only may ignore the uncertain factors and lead project managers (PMs) to wrong decision making. Prediction intervals (PIs) with confidence levels (CLs) present a more reasonable representation of reality, potentially helping PMs to make better-informed decisions and enable more flexibility in these decisions. However, existing methods for PIs either have strong limitations or are unable to provide informative PIs. To develop a “better” effort predictor, we propose a novel PI estimator called Synthetic Bootstrap ensemble of Relevance Vector Machines (SynB-RVM) that adopts Bootstrap resampling to produce multiple RVM models based on modified training bags whose replicated data projects are replaced by their synthetic counterparts. We then provide three ways to assemble those RVM models into a final probabilistic effort predictor, from which PIs with CLs can be generated. When used as a point estimator, SynB-RVM can either significantly outperform or have similar performance compared with other investigated methods. When used as an uncertain predictor, SynB-RVM can achieve significantly narrower PIs compared to its base learner RVM. Its hit rates and relative widths are no worse than the other compared methods that can provide uncertain estimation.

Convex formulation of multiple instance learning from positive and unlabeled bags

Multiple instance learning (MIL) is a variation of traditional supervised learning problems where data (referred to as bags) are composed of sub-elements (referred to as instances) and only bag labels are available. MIL has a variety of applications such as content-based image retrieval, text categorization, and medical diagnosis. Most of the previous work for MIL assume that training bags are fully labeled. However, it is often difficult to obtain an enough number of labeled bags in practical situations, while many unlabeled bags are available. A learning framework called PU classification (positive and unlabeled classification) can address this problem. In this paper, we propose a convex PU classification method to solve an MIL problem. We experimentally show that the proposed method achieves better performance with significantly lower computation costs than an existing method for PU-MIL.

ISBDD Model for Classification of Hyperspectral Remote Sensing Imagery.

The diverse density (DD) algorithm was proposed to handle the problem of low classification accuracy when training samples contain interference such as mixed pixels. The DD algorithm can learn a feature vector from training bags, which comprise instances (pixels). However, the feature vector learned by the DD algorithm cannot always effectively represent one type of ground cover. To handle this problem, an instance space-based diverse density (ISBDD) model that employs a novel training strategy is proposed in this paper. In the ISBDD model, DD values of each pixel are computed instead of learning a feature vector, and as a result, the pixel can be classified according to its DD values. Airborne hyperspectral data collected by the Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) sensor and the Push-broom Hyperspectral Imager (PHI) are applied to evaluate the performance of the proposed model. Results show that the overall classification accuracy of ISBDD model on the AVIRIS and PHI images is up to 97.65% and 89.02%, respectively, while the kappa coefficient is up to 0.97 and 0.88, respectively.

Incorporating Diversity and Informativeness in Multiple-Instance Active Learning

Multiple-instance active learning (MIAL) is a paradigm to collect sufficient training bags for a multiple-instance learning (MIL) problem, by selecting and querying the most valuable unlabeled bags iteratively. Existing works on MIAL evaluate an unlabeled bag by its informativeness with regard to the current classifier, but neglect the internal distribution of its instances, which can reflect the diversity of the bag. In this paper, two diversity criteria, i.e., clustering-based diversity and fuzzy rough set based diversity, are proposed for MIAL by utilizing a support vector machine (SVM) based MIL classifier. In the first criterion, a kernel $k$ -means clustering algorithm is used to explore the hidden structure of the instances in the feature space of the SVM, and the diversity degree of an unlabeled bag is measured by the number of unique clusters covered by the bag. In the second criterion, the lower approximations in fuzzy rough sets are used to define a new concept named dissimilarity degree, which depicts the uniqueness of an instance so as to measure the diversity degree of a bag. By incorporating the proposed diversity criteria with existing informativeness measurements, new MIAL algorithms are developed, which can select bags with both high informativeness and diversity. Experimental comparisons demonstrate the feasibility and effectiveness of the proposed methods.

Effectiveness of the training courses of kindergarten teachers in Riyadh in the development of their professional qualifications: فاعلية دورات مركز تدريب معلمات رياض الأطفال بمدينة الرياض في تنمية كفايتهن المهنية

The aim of this study was to investigate the effectiveness of the training courses of kindergarten teachers. The study used the descriptive method. The target population of the study was 46 trainees from the training center of kindergartens, teachers and kindergartens in Riyadh. In this study, three tools were used to collect the information, namely, the questionnaire and the training impact form within the training bag of the center and the interview. The study also presented four questions. This study draws on the importance of the local and international approach to the issues of early childhood and the associated educational programs and the training of teachers. And the importance of the educational process, which is the teacher one of the elements or pillars of effective and the importance of the early stage of education as it is the basic stage in which the formation of knowledge and information of the child.Also contribute to the development and renewal of the teacher's experiences through training programs both during the service or rehabilitation programs and programs aimed at developing and renewing the teacher's experiences . It also helps to compensate for the lack of studies and researches related to training programs and the measurement of their results, especially those programs provided to kindergarten teachers. The main findings of the study are that there is a correlation between the training program and the future of the trainees. And contributes to the identification of the characteristics of growth in kindergartens where the average arithmetic (3). The program offers suggestions to deal with the behavioral class problems of the child and contribute to the identification of ways to observe the behavior and modify it in addition to the administrative supervision p To the course contributed to the achievement of its objectives, where the average arithmetic mean (2.93). The results showed high approval of the planning axis of the educational unit with an average of (4.09) and high approval of the axis of dealing with children with an average of (4.29). As for the most important obstacles, it was easy to reach the training center where it reached the average of (2.26) and the candidates for training on the training program well ahead of time with an average of (2.30). The study recommended ways to increase the effectiveness of the training center for kindergartens teachers and work to remove the obstacles that limit the enrollment of private sector teachers such courses ..

Salient Object Detection via Multiple Instance Learning.

Object proposals are a series of candidate segments containing objects of interest, which are taken as preprocessing and widely applied in various vision tasks. However, most of existing saliency approaches only utilize the proposals to compute a location prior. In this paper, we naturally take the proposals as the bags of instances of multiple instance learning (MIL), where the instances are the superpixels contained in the proposals, and formulate saliency detection problem as a MIL task (i.e., predict the labels of instances using the classifier in the MIL framework). This method allows some flexibility in finding a decision boundary based on the bag-level representations and can identify salient superpixels from ambiguous proposals. In addition, we introduce the MIL to an optimization mechanism, which iteratively updates training bags from easy to complex ones to learn a strong model. The significant improvement can be consistently achieved when applying the optimization model to existing saliency approaches. Extensive experiments demonstrate that the proposed algorithms perform favorably against the stateof- art saliency detection methods on several benchmark datasets.

Heart beat characterization from ballistocardiogram signals using extended functions of multiple instances.

A multiple instance learning (MIL) method, extended Function of Multiple Instances (eFUMI), is applied to ballistocardiogram (BCG) signals produced by a hydraulic bed sensor. The goal of this approach is to learn a personalized heartbeat "concept" for an individual. This heartbeat concept is a prototype (or "signature") that characterizes the heartbeat pattern for an individual in ballistocardiogram data. The eFUMI method models the problem of learning a heartbeat concept from a BCG signal as a MIL problem. This approach elegantly addresses the uncertainty inherent in a BCG signal (e. g., misalignment between training data and ground truth, mis-collection of heartbeat by some transducers, etc.). Given a BCG training signal coupled with a ground truth signal (e.g., a pulse finger sensor), training "bags" labeled with only binary labels denoting if a training bag contains a heartbeat signal or not can be generated. Then, using these bags, eFUMI learns a personalized concept of heartbeat for a subject as well as several non-heartbeat background concepts. After learning the heartbeat concept, heartbeat detection and heart rate estimation can be applied to test data. Experimental results show that the estimated heartbeat concept found by eFUMI is more representative and a more discriminative prototype of the heartbeat signals than those found by comparison MIL methods in the literature.

Fetal Ultrasound Image Classification Using a Bag-of-words Model Trained on Sonographers’ Eye Movements

The acquisition of fetal biometric measurements via 2-D B-Mode ultrasound (US) scans is crucial for fetal monitoring. However, acquiring standardised head, abdominal and femoral image planes is challenging due to variable image quality. There remains a significant discrepancy between the way automated computer vision algorithms and human sonographers perform this task; this paper contributes to the attempt to bridge this gap by building knowledge of US image perception into a pipeline for classifying images obtained during 2-D fetal US scans. We record the eye movements of 10 participants performing 4 2-D US scans each, on a phantom fetal model at varying orientations. We analyse their eye movements to establish which high-level constraints and visual cues are used to localise the standardised abdominal plane. We then build a vocabulary of visual words trained on SURF descriptors extracted around eye fixations, and use the resulting bag of words model to classify head, abdominal and femoral image frames acquired during 10 clinical US scans and 10 further phantom US scans. On phantom data, we achieve classification accuracies of 89%, 87% and 85% for the head, abdominal and femoral images respectively. On clinical data, we achieve classification accuracies of 76%, 68% and 64% for the head, abdominal and femoral images respectively. This constitutes the first insight into image perception during real time US scanning, and a proof of concept for training bag of words models for US image analysis on human eye movements.

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.

فـاعـلـیـة بـرنـامـج تدریـبـی مـقـتـرح فـی إکـسـاب مـعـلـمـات الـریـاضـیـات بـعـض مـهـارات الـتـعـلـم الـنـشـط وخـفـض قـلـق الـریـاضـیـات لــدى طـالـبـاتـهـن

هدفالبحثإلىمعرفةفاعليةبرنامجتدريبيمقترحفيإکسابمعلمات الرياضياتبعضمهارات التعلمالنشط وفي خفض قلق الرياضيات لدى طالبتهن ،وتحقيقالهذاالهدفتماستخدامالمنهج التجريبيحيثطبق البحثعلىعينتين: العينةالأولىمنمعلماتالرياضياتبالمرحلةالمتوسطةبمکتب التربيةوالتعليمبمدينة تبوک ،بلغعددهم(11)معلمةقدملهمبرنامجاتدريبياحولالتعلمالنشطوتمقياسأدائهم القبليوالبعديبواسطةمقياسالأداءلمهاراتالتعلمالنشط ،والعينةالثانية طالبات المعلمات الذينحضروا البرنامجالتدريبيمنطالبات المرحلة المتوسطة ،وقدبلغعددهم(150)طالبةتمتدريسهمبالتعلم النشط ،وتمقياسأدائهمعلى مقياس القلق الرياضيقبلياوبعديا ،تماستخدام اختبار ت للمجموعاتالمترابطة ،لاختبارفروض البحث ،وقدتوصل البحثإلىالنتائجالتالية: ١ -توجدفروقذاتدلالةإحصائيةعندمستوى = α )0.05) بين متوسطي أداء معلمات الرياضيات (عينة البحث )علىمقياسالأداءلمهاراتالتعلمالنشطقبلتعرضهمللبرنامجالتدريبيالمقترح،وبعدهلصالح التطبيقالبعدي. 2ـ انخفاض مستوى القلق الرياضي لدى طالبات المرحلة المتوسطة والذين درسوا باستخدام التعلم النشط . وفيضوءنتائجالبحثتمتقديمعددمنالتوصياتمن اهمها : 1ـ تدريب معلمي الرياضيات أثناء الخدمة على إستراتيجيات التعلم النشط المختلفة , لما لها من أثر إيجابي على إکسابهم مهارات التعلم النشط المختلفة مثل (فکر – زاوج – شارک , التدريس التبادلي , حل المشکلات) وما سوف ينعکس على خفض مستوى القلق الرياضي لدى الطلاب. 2ـ تزويد المعلمين بحقائب تدريبية تحتوي على البرامج التدريبية بأدبياتها وبياناتها وحقائقها وموادها التدريبية ؛ مما يوفر للمعلمين ما يرجعون إليه لتأکيد الخبرات والتوسع فيها وتطويرها. 3ـ الترکيز على کيفية تنظيم البيئة التعليمية في فصول الرياضيات , بحيث يتم التعلم في جو من الإثارة والتشويق , من خلال الاهتمام بصياغة مشکلات رياضية فعالة , تمس واقع الطلاب , وتسهم في خفض القلق الرياضي للطلاب. The research title the actuating of an open training programmes was suggested is giving the mathematics teacher some of the educational learning and decreasing the math. many to their students Dr. sherine Salah Abdulhakeem Ahmad. Asst . Prof of methodology of math and curriculum at faculty of girsl , Ain Shams University. The research goal to know the utility if suggested having program in giving the math. Teachers some active learning talents and in decreasing the math warry to this students and to achieve that goal the trial contetnt was used and it was applied on two samples: the first one from the math teacher at intermediate stage at the education office in tabuk city. With number equal (11) teacher they were offered a training programme in itnitial learning and their befor and after performance was measured by the programme math for talent of initial programme. The second sample The student of female teachers who attented the training programme it the student of intermediate stage. Their number was (150 ) student were teached by the initial learning and their results was measured by the worry of math before and after and then using the (t) test for gatherd groups . the research lypostucs and the research vouched to the following results: 1- There are different statically at the level of (α= 0.05) between the average of math teachers performance panmeters (sample) by vital learning talents before facing the suggested training programme and distance for the after application> 2- The decrease of the math worry level at female students of the intermediate stage who studied by using the vital learning. And According to the research result there were some recommendations: 1- Training of mathematics teachers during the service on active learning different strategies . as they have a positive impact on them acquire skills such as active learning different ( think – pair- participated, reciprocal, teaching proplem solving ) and will reflected in reducing the level if anxiety sports among students. 2- Provide teachers with training bags containing bodbeadtha training programs , data and realities and training materials ; giving teachers what they refer to experiences to confirm and expand and develpoe. 3- Focus on how to organize the learning environment in mathematics classes. So learning in atmosphere of excitement and thrill, through attention to the formulation of effective mathematical problems, affecting the reality of the students, and contribute to the reduction of math anxiety for students.

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.

Multiple-instance learning with pairwise instance similarity

Abstract Multiple-Instance Learning (MIL) has attracted much attention of the machine learning community in recent years and many real-world applications have been successfully formulated as MIL problems. Over the past few years, several Instance Selection-based MIL (ISMIL) algorithms have been presented by using the concept of the embedding space. Although they delivered very promising performance, they often require long computation times for instance selection, leading to a low efficiency of the whole learning process. In this paper, we propose a simple and efficient ISMIL algorithm based on the similarity of pairwise instances within a bag. The basic idea is selecting from every training bag a pair of the most similar instances as instance prototypes and then mapping training bags into the embedding space that is constructed from all the instance prototypes. Thus, the MIL problem can be solved with the standard supervised learning techniques, such as support vector machines. Experiments show that the proposed algorithm is more efficient than its competitors and highly comparable with them in terms of classification accuracy. Moreover, the testing of noise sensitivity demonstrates that our MIL algorithm is very robust to labeling noise