Multi-Instance Nonparallel Tube Learning.

In multi-instance learning, each example is represented by a bag of instances while associated with a binary label. Under standard multi-instance learning settings, one example is labeled as a positive bag if at least one of its instances is positive. Otherwise, it is labeled as a negative bag. Although based on the above assumption, standard multi-instance learning has achieved much success in solving diverse learning tasks, there are still many real-world problems where this assumption may not necessarily hold. Therefore, researchers aimed to expand the underlying assumption of standard multi-instance learning where two frameworks of generalized multi-instance learning have been proposed. In this paper, the problem definition, learning algorithms and also experimental data sets related to either generalized multi-instance learning framework are briefly reviewed.

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.

Multi-instance dictionary learning via multivariate performance measure optimization

Multi-instance learning was first proposed by Dietterich et al. (Artificial Intelligence 89(1–2):31–71, 1997) when they were investigating the problem of drug activity prediction. Here, the training set is composed of labeled bags, each of which consists of many unlabeled instances. And the goal of this learning framework is to learn some classifier from the training set for correctly labeling unseen bags. After Dietterich et al., many studies about this new learning framework have been started and many new algorithms have been proposed, for example, DD, EM-DD, Citation-kNN and so on. All of these algorithms are working on the full data set. But as in single-instance learning, different feature in training set has different effect on the training about classifier. In this paper, we will study the problem about feature selection in multi-instance learning. We will extend the data reliability measure and make it select the key feature in multi-instance scenario.

In this paper, a multi-instance learning based CBIR (content-based image retrieval) approach is presented, and multi-instance learning method is applied in CBIR, in order to deal with the inherent ambiguity of images. First of all the whole image is regards as a multi-instance bag, secondly the image is partitioned into multi-regions by using adaptive k-means image segmentation method, and then query images posed by the user are transformed into corresponding positive and negative bags and a EM-DD(expectation maximization diverse density) algorithm is employed for image retrieval and relevance feedback. Finally, it makes the users get satisfying result.

Protein localization information is essential for understanding protein functions and their roles in various biological processes. The image-based prediction methods of protein subcellular localization have emerged in recent years because of the advantages of microscopic images in revealing spatial expression and distribution of proteins in cells. However, the image-based prediction is a very challenging task, due to the multi-instance nature of the task and low quality of images. In this paper, we propose a multi-task learning strategy and mask generation to enhance the prediction performance. Furthermore, we also investigate effective multi-instance learning schemes. We collect a large-scale dataset from the Human Protein Atlas database, and the experimental results show that the proposed multi-task multi-instance learning model outperforms both single-instance learning and common multi-instance learning methods by large margins.

Twin support vector machine (TSVM) has been widely applied to classification problems. But TSVM is sensitive to outliers and is not efficient enough to realize feature selection. To overcome the shortcomings of TSVM, we propose a novel sparse twin support vector machine with the correntropy-induced loss (C-STSVM), which is inspired by the robustness of the correntropy-induced loss and the sparsity of the \(\ell _1\)-norm regularization. The objective function of C-STSVM includes the correntropy-induced loss that replaces the hinge loss, and the \(\ell _1\)-norm regularization that can make the decision model sparse to realize feature selection. Experiments on real-world datasets with label noise and noise features demonstrate the effectiveness of C-STSVM in classification accuracy and confirm the above conclusion further.

Twin support vector machine (TWSVM), which solves classification problems through constructing two nonparallel planes by solving two related SVM-type problems, has become a hot spot in the field of machine learning. In addition to keeping the advantages of SVM, the classification performance of TWSVM is also significantly better than that of SVM. However, there are at least two parameters in TWSVM which need to specify. These parameters make great difference to the classification result of TWSVM and to find the optimal parameters is an important but difficult work. In order to overcome this deficiency, in this paper, we propose the twin support vector machines based on Fruit Fly Optimization Algorithm (FOA-TWSVM). This algorithm uses FOA-TWSVM, which has high optimization ability and small computation complexity, to select the parameters for TWSVM. The experimental results show that this algorithm is able to find the suitable parameters, and has higher classification accuracy compared with some other algorithms.

Multi-instance Deep Learning of Ultrasound Imaging Data for Pattern Classification of Congenital Abnormalities of the Kidney and Urinary Tract in Children

Over the past decades, 1-norm techniques based on algorithms are widely used to suppress input features. Quite different from traditional 1-norm support vector machine (SVM), direct 1-norm optimization based on the primal problem of nonparallel plane classifiers like generalized proximal support vector machine, twin support vector machine (TWSVM) and least squares twin support vector machine (LSTSVM) are not capable of generating very sparse solutions that are vital for classification and can make them easier to store and faster to compute. To address the issue, in this paper, we develop a feature selection method for LSTSVM, called a feature selection method for nonparallel plane support vector machine classification (FLSTSVM), which is specially designed for strong feature suppression. We incorporate a Tikhonov regularization term to the objective of LSTSVM, and then minimize its 1-norm measure. Solution of FLSTSVM can follow directly from solving two smaller quadratic programming problems (QPPs) arising from two primal QPPs as opposed to two dual ones in TWSVM. FLSTSVM is capable of generating very sparse solutions. This means that FLSTSVM can reduce input features, for the linear case. When a nonlinear classifier is used, few kernel functions determine the classifier. In addition to having strong feature suppression, the edge of our method still lies in its faster computing time compared to that of TWSVM, Newton Method for Linear Programming SVM (NLPSVM) and LPNewton. Lastly, this algorithm is compared on public data sets, as well as an Exclusive Or (XOR) example.

User attribute is an essential factor for personalized recommendation and targeted advertising. Therefore, there have been a number of studies to identify user attributes automatically from SNS postings, since the postings reveal various attributes of writers. Many kinds of machine learning methods have been applied to automatic identification of user attributes as a candidate solution, but they suffer from two major problems. First, there are many postings in SNS that do not deliver any information about writers. Then, learning from SNS postings results in a biased model by these irrelevant postings. Second, the postings of a SNS user are somewhat related one another. However, most machine learning methods ignore this information, since they assume that data are independently and identically distributed. In order to solve these problems in user attribute identification, this paper proposes a novel method based on non-i.i.d. multi-instance learning. Since multi-instance learning treats all postings by a user as a bag and learns user attribute identification with such bags, not with postings, the first problem is solved. In addition, the proposed method assumes that the postings by a single user have a structure. By incorporating this assumption into the multi-instance learning, the second problem is solved. Our experimental results show that consideration of these two problems in automatic user attribute identification results in performance improvement.

Glaucoma is a group of eye diseases that cause an irreparable decrease in the view field which is the second most common and leading causes of blindness among retinal diseases. Computer aided diagnosis (CAD) system is an emerging field in medical informatics which has high importance for providing prognosis of diseases. Research efforts have reported with increasing confirmation that the twin support vector machines (TWSVM) have greater accurate diagnosis ability. The goal of TWSVM is to construct two non-parallel planes for each class such that each hyper-plane is closer to one of two classes and as far as possible from the other one. In this paper, we propose a new CAD system for glaucoma diagnosis using TWSVM and three heterogeneous families of feature extraction. In this work, we have used 169 images to classify into normal and glaucoma classes. The performance of the method is evaluated using classification accuracy, sensitivity, specificity and receiver operating characteristic (ROC) curves. The results show that the highest classification accuracy (98,53%) is obtained for the TWSVM using Gaussian kernel function, and this is very promising compared to the SVM results.

To develop a multi-modal deep learning method for automatic classification of immune-mediated glomerular diseases based on images of optical microscopy (OM), immunofluorescence microscopy (IM), and transmission electron microscopy (TEM). We retrospectively collected the pathological images from 273 patients and constructed a multi-modal multi- instance model for classification of 3 immune-mediated glomerular diseases, namely immunoglobulin A nephropathy (IgAN), membranous nephropathy (MN), and lupus nephritis (LN). This model adopts an instance-level multi-instance learning (I-MIL) method to select the TEM images for multi-modal feature fusion with the OM images and IM images of the same patient. By comparing this model with unimodal and bimodal models, we explored different combinations of the 3 modalities and the optimal methods for modal feature fusion. The multi-modal multi-instance model combining OM, IM, and TEM images had a disease classification accuracy of (88.34±2.12)%, superior to that of the optimal unimodal model [(87.08±4.25)%] and that of the optimal bimodal model [(87.92±3.06)%]. This multi- modal multi- instance model based on OM, IM, and TEM images can achieve automatic classification of immune-mediated glomerular diseases with a good classification accuracy.

Weighted Twin Support Vector Machines with Local Information and its application

Multi-instance learning is a type of weakly supervised learning. It deals with tasks where the data is a set of bags and each bag is a set of instances. Only the bag labels are observed whereas the labels for the instances are unknown. An important advantage of multi-instance learning is that by representing objects as a bag of instances, it is able to preserve the inherent dependencies among parts of the objects. Unfortunately, most existing algorithms assume all instances to be identically and independently distributed, which violates real-world scenarios since the instances within a bag are rarely independent. In this work, we propose the Multi-Instance Variational Autoencoder (MIVAE) algorithm which explicitly models the dependencies among the instances for predicting both bag labels and instance labels. Experimental results on several multi-instance benchmarks and end-to-end medical imaging datasets demonstrate that MIVAE performs better than state-of-the-art algorithms for both instance label and bag label prediction tasks.