This paper tackles the problem of the worst-class error rate, instead of the standard error rate averaged over all classes. For example, a three-class classification task with class-wise error rates of 10%, 10%, and 40% has a worst-class error rate of 40%, whereas the average is 20% under the class-balanced condition. The worst-class error is important in many applications. For example, in a medical image classification task, it would not be acceptable for the malignant tumor class to have a 40% error rate, while the benign and healthy classes have a 10% error rates. To avoid overfitting in worst-class error minimization using Deep Neural Networks (DNNs), we design a problem formulation for bounding the worst-class error instead of achieving zero worst-class error. Moreover, to correctly bound the worst-class error, we propose a boosting approach which ensembles DNNs. We give training and generalization worst-class-error bound. Experimental results show that the algorithm lowers worst-class test error rates while avoiding overfitting to the training set.

Patient-level diagnosis of severity in ulcerative colitis (UC) is common in clinical practice, where the most severe score for a patient is typically recorded as the diagnosis result. However, previous UC classification methods (i.e., image-level estimation) mainly assumed the input was a single image. Thus, these methods can not utilize severity labels recorded in real clinical settings. In this paper, we propose a patient-level severity estimation method by a transformer with selective aggregator tokens, where a severity label is estimated from multiple images taken from a patient, similar to a clinical setting. Our method can effectively aggregate features of severe parts from a set of images captured in each patient, and it facilitates improving the discriminative ability between adjacent severity classes. Experiments demonstrate the effectiveness of the proposed method on two datasets compared with the state-of-the-art MIL methods. Moreover, we evaluated our method using real clinical data and confirmed that our method outperformed the previous image-level methods. The code is publicly available at https://github.com/Shiku-Kaito/Ordinal-Multiple-instance-Learning-for-Ulcerative-Colitis-Severity-Estimation.

Learning from label proportions (LLP) is a kind of weakly supervised learning that trains an instance-level classifier from label proportions of bags, which consist of sets of instances without using instance labels. A challenge in LLP arises when the number of instances in a bag (bag size) is numerous, making the traditional LLP methods difficult due to GPU memory limitations. This study aims to develop an LLP method capable of learning from bags with large sizes. In our method, smaller bags (mini-bags) are generated by sampling instances from large-sized bags (original bags), and these mini-bags are used in place of the original bags. However, the proportion of a mini-bag is unknown and differs from that of the original bag, leading to overfitting. To address this issue, we propose a perturbation method for the proportion labels of sampled mini-bags to mitigate overfitting to noisy label proportions. This perturbation is added based on the multivariate hypergeometric distribution, which is statistically modeled. Additionally, loss weighting is implemented to reduce the negative impact of proportions sampled from the tail of the distribution. Experimental results demonstrate that the proportion label perturbation and loss weighting achieve classification accuracy comparable to that obtained without sampling. Our codes are available at https://github.com/stainlessnight/LLP-LargeBags.

The paper proposes a novel problem in multi-class Multiple-Instance Learning (MIL) called Learning from the Majority Label (LML). In LML, the majority class of instances in a bag is assigned as the bag’s label. LML aims to classify instances using bag-level majority classes. This problem is valuable in various applications. Existing MIL methods are unsuitable for LML due to aggregating confidences, which may lead to inconsistency between the bag-level label and the label obtained by counting the number of instances for each class. This may lead to incorrect instance-level classification. We propose a counting network trained to produce the bag-level majority labels estimated by counting the number of instances for each class. This led to the consistency of the majority class between the network outputs and one obtained by counting the number of instances. Experimental results show that our counting network outperforms conventional MIL methods on four datasets1.

This paper proposes a novel and efficient method for Learning from Label Proportions (LLP), whose goal is to train a classifier only by using the class label proportions of instance sets, called bags. We propose a novel LLP method based on an online pseudo-labeling method with regret minimization. As opposed to the previous LLP methods, the proposed method effectively works even if the bag sizes are large. We demonstrate the effectiveness of the proposed method using some benchmark datasets.

Learning from label proportions (LLP) is a promising weakly supervised learning problem. In LLP, a set of instances (bag) has label proportions, but no instance-level labels are given. LLP aims to train an instance-level classifier by using the label proportions of the bag. In this paper, we propose a bag-level data augmentation method for LLP called MixBag, based on the key observation from our preliminary experiments; that the instance-level classification accuracy improves as the number of labeled bags increases even though the total number of instances is fixed. We also propose a confidence interval loss designed based on statistical theory to use the augmented bags effectively. To the best of our knowledge, this is the first attempt to propose bag-level data augmentation for LLP. The advantage of MixBag is that it can be applied to instance-level data augmentation techniques and any LLP method that uses the proportion loss. Experimental results demonstrate this advantage and the effectiveness of our method.

In statistical learning, many problem formulations have been proposed so far, such as multi-class learning, complementarily labeled learning, multi-label learning, multi-task learning, which provide theoretical models for various real-world tasks. Although they have been extensively studied, the relationship among them has not been fully investigated. In this work, we focus on a particular problem formulation called Multiple-Instance Learning (MIL), and show that various learning problems including all the problems mentioned above with some of new problems can be reduced to MIL with theoretically guaranteed generalization bounds, where the reductions are established under a new reduction scheme we provide as a by-product. The results imply that the MIL-reduction gives a simplified and unified framework for designing and analyzing algorithms for various learning problems. Moreover, we show that the MIL-reduction framework can be kernelized.

We propose an online visual-object tracking method that is robust even in an adversarial environment, where various disturbances may occur on the target appearance, etc. The proposed method is based on a delayed-Hedge algorithm for aggregating multiple arbitrary online trackers with adaptive weights. The robustness in the tracking performance is guaranteed theoretically in term of "regret" by the property of the delayed-Hedge algorithm. Roughly speaking, the proposed method can achieve a similar tracking performance as the best one among all the trackers to be aggregated in an adversarial environment. The experimental study on various tracking tasks shows that the proposed method could achieve state-of-the-art performance by aggregating various online trackers.

We propose a new formulation of multiple-instance learning (MIL), in which a unit of data consists of a set of instances called a bag. The goal is to find a good classifier of bags based on the similarity with a "shapelet" (or pattern), where the similarity of a bag with a shapelet is the maximum similarity of instances in the bag. In previous work, some of the training instances have been chosen as shapelets with no theoretical justification. In our formulation, we use all possible, and thus infinitely many, shapelets, resulting in a richer class of classifiers. We show that the formulation is tractable, that is, it can be reduced through linear programming boosting (LPBoost) to difference of convex (DC) programs of finite (actually polynomial) size. Our theoretical result also gives justification to the heuristics of some previous work. The time complexity of the proposed algorithm highly depends on the size of the set of all instances in the training sample. To apply to the data containing a large number of instances, we also propose a heuristic option of the algorithm without the loss of the theoretical guarantee. Our empirical study demonstrates that our algorithm uniformly works for shapelet learning tasks on time-series classification and various MIL tasks with comparable accuracy to the existing methods. Moreover, we show that the proposed heuristics allow us to achieve the result in reasonable computational time.

In this paper, we analyze logo designs by using machine learning, as a promising trial of graphic design analysis. Specifically, we will focus on favicon images, which are tiny logos used as company icons on web browsers, and analyze them to understand their trends in individual industry classes. For example, if we can catch the subtle trends in favicons of financial companies, they will suggest to us how professional designers express the atmosphere of financial companies graphically. For the purpose, we will use top-rank learning, which is one of the recent machine learning methods for ranking and very suitable for revealing the subtle trends in graphic designs.

Signature verification systems suffer from imbalanced learning, which imposes strict requirements on classifiers. The standard classification approaches, such as SVM, often degrade the performance for imbalanced data or require additional parameters for data balancing. In this study, as a new approach for signature verification, we use RankSVM as the writer-dependent classifiers, which theoretically guarantees the generalization performance for imbalanced data. To investigate the ability of RankSVM for solving imbalanced learning problems in signature verification tasks, the extensive experiments are conducted on bitmaps of GPDS-150, GPDS-300, GPDS-600, and GPDS-1000 datasets and deep features of GPDS-960 dataset. The experimental results demonstrate that the RankSVM-based approach obtains a nearly equivalent performance with the state-of-the-art method on deep features of the GPDS-960 dataset, and achieves significantly better performance than standard-SVM-based approach on bitmaps of GPDS-150, GPDS-300, GPDS-600, and GPDS-1000 datasets.

To discover teaching knowledge efficiently, we must extract the various teaching activities from educational data. In this paper, through the use of e-book logs and techniques of time-series analysis, we describe a method of practicing teaching analytics in face-to-face classes, one which enable us to extract the teaching activity efficiently and accurately.

Students' reflective writings are useful not only for students themselves but also teachers. It is important for teachers to know which concepts were understood well by students and which concepts were not, to continuously improve their classes. However, it is difficult for teachers to thoroughly read the journals of more than one hundred students. In this paper, we propose a novel method to extract common topics and students' common impressions against them from students' journals. Weekly keywords are discovered from journals by scoring noun words with a measure based on TF-IDF term weighting scheme, and then we analyze co-occurrence relationships between extracted keywords and adjectives. We employs nonnegative matrix factorization, one of the topic modeling techniques, to discover the hidden impression topics from the co-occurrence relationships. As a case study, we applied our method on students' journals of the course "Information Science" held in our university. Our experimental results show that conceptual keywords are successfully extracted, and four significant impression topics are identified. We conclude that our analysis method can be used to collectively understand the impressions of students from journal texts.

Many universities choose the C programming language (C) as the first one they teach their students, early on in their program. However, students often consider programming courses difficult, and these courses often have among the highest dropout rates of computer science courses offered. It is therefore critical to provide more effective instruction to help students understand the syntax of C and prevent them losing interest in programming. In addition, homework and paper-based exams are still the main assessment methods in the majority of classrooms. It is difficult for teachers to grasp students' learning situation due to the large amount of evaluation work. To facilitate teaching and learning of C, in this article we propose a system---LAPLE (Learning Analytics in Programming Language Education)---that provides a learning dashboard to capture the behavior of students in the classroom and identify the different difficulties faced by different students looking at different knowledge. With LAPLE, teachers may better grasp students' learning situation in real time and better improve educational materials using analysis results. For their part, novice undergraduate programmers may use LAPLE to locate syntax errors in C and get recommendations from educational materials on how to fix them.