Learning Classifiers on Positive and Unlabeled Data with Policy Gradient

Improving state-of-health estimation for lithium-ion batteries via unlabeled charging data

We consider the problem of adding a large unlabeled sample from the target domain to boost the performance of a domain adaptation algorithm when only a small set of labeled examples is available from the target domain. In particular, we consider the problem setting motivated by the tasks of splice site prediction and protein localization. For example, for splice site prediction, annotating a genome using machine learning requires a lot of labeled data, whereas for non-model organisms, there are only some labeled data and lots of unlabeled data. With domain adaptation one can leverage the large amount of data from a related model organism, along with the labeled and unlabeled data from the organism of interest to train a classifier for the latter. Our goal is to analyze the three approaches of incorporating the unlabeled data—with soft labels only (i.e., Expectation-Maximization), with hard labels only (i.e., self-training), or with both soft and hard labels—for the splice site prediction and protein localization in particular, and more broadly for a general iterative domain adaptation setting. We provide empirical results on splice site prediction and protein localization indicating that using a combination of soft and hard labels performs as good as the best of the other two approaches of integrating unlabeled data.

Active Learning is a machine learning and data mining technique that selects the most informative samples for labeling and uses them as training data; it is especially useful when there are large amount of unlabeled data and labeling them is expensive. Recently, batch-mode active learning, where a set of samples are selected concurrently for labeling, based on their collective merit, has attracted a lot of attention. The objective of batch-mode active learning is to select a set of informative samples so that a classifier learned on these samples has good generalization performance on the unlabeled data. Most of the existing batch-mode active learning methodologies try to achieve this by selecting samples based on certain criteria. In this article we propose a novel criterion which achieves good generalization performance of a classifier by specifically selecting a set of query samples that minimize the difference in distribution between the labeled and the unlabeled data, after annotation. We explicitly measure this difference based on all candidate subsets of the unlabeled data and select the best subset. The proposed objective is an NP-hard integer programming optimization problem. We provide two optimization techniques to solve this problem. In the first one, the problem is transformed into a convex quadratic programming problem and in the second method the problem is transformed into a linear programming problem. Our empirical studies using publicly available UCI datasets and two biomedical image databases demonstrate the effectiveness of the proposed approach in comparison with the state-of-the-art batch-mode active learning methods. We also present two extensions of the proposed approach, which incorporate uncertainty of the predicted labels of the unlabeled data and transfer learning in the proposed formulation. In addition, we present a joint optimization framework for performing both transfer and active learning simultaneously unlike the existing approaches of learning in two separate stages, that is, typically, transfer learning followed by active learning. We specifically minimize a common objective of reducing distribution difference between the domain adapted source, the queried and labeled samples and the rest of the unlabeled target domain data. Our empirical studies on two biomedical image databases and on a publicly available 20 Newsgroups dataset show that incorporation of uncertainty information and transfer learning further improves the performance of the proposed active learning based classifier. Our empirical studies also show that the proposed transfer-active method based on the joint optimization framework performs significantly better than a framework which implements transfer and active learning in two separate stages.

In positive and unlabeled (PU) learning problems, only positive examples are labeled. Unlabeled data contain both positive and negative examples. Studies show that positive examples of (secondary) diagnoses, and clinical conditions, such as sepsis, are present in unlabeled hospital administrative data, potentially distorting hospital reimbursement systems, and negatively affecting hospitals' revenue and profitability. We investigate whether PU learning is suitable for improving the quality of hospital administrative data. We train three models on 313,434 hospital cases using hospital cost features: two based on the two-step "spy" approach and one using a robust PU learning method. For model evaluation, we rely exclusively on positive examples due to the PU setting. To further assess model performance, we perform an external validity check: We relabel unlabeled sepsis cases, derive new sepsis rates, and compare them to those reported in medical record review studies. All models identify true positives well in unseen data. External validity checks show, however, that only the robust PU learner effectively discriminates between positives and negatives in the unlabeled data, yielding new sepsis rates within the range of sepsis rates reported in medical record review studies. PU learning can improve the quality of hospital administrative data, but its effectiveness depends strongly on the choice of learning approach and classifier. The output of a PU learner can potentially improve hospital reimbursement systems, hospital revenue and profitability management, and sensitivity analyses in healthcare management science, health economics, health services research, and disease surveillance.

Exploiting unlabeled data for battery state-of-health estimation using transformer-LSTM neural network with semi-supervised learning

While recent studies on semi-supervised learning have shown remarkable progress in leveraging both labeled and unlabeled data, most of them presume a basic setting of the model is randomly initialized. In this work, we consider semi-supervised learning and transfer learning jointly, leading to a more practical and competitive paradigm that can utilize both powerful pre-trained models from source domain as well as labeled/unlabeled data in the target domain. To better exploit the value of both pre-trained weights and unlabeled target examples, we introduce adaptive consistency regularization that consists of two complementary components: Adaptive Knowledge Consistency (AKC) on the examples between the source and target model, and Adaptive Representation Consistency (ARC) on the target model between labeled and unlabeled examples. Examples involved in the consistency regularization are adaptively selected according to their potential contributions to the target task. We conduct extensive experiments on popular benchmarks including CIFAR-10, CUB-200, and MURA, by fine-tuning the ImageNet pre-trained ResNet-50 model. Results show that our proposed adaptive consistency regularization outperforms state-of-the-art semi-supervised learning techniques such as Pseudo Label, Mean Teacher, and FixMatch. Moreover, our algorithm is orthogonal to existing methods and thus able to gain additional improvements on top of MixMatch and FixMatch. Our code is available at https://github.com/Walleclipse/Semi-Supervised-Transfer-Learning-Paddle.

This paper studies the problem of Positive Unlabeled learning (PU learning), where positive and unlabeled examples are used for training. Naive Bayes (NB) and Tree Augmented Naive Bayes (TAN) have been extended to PU learning algorithms (PNB and PTAN). However, they require user-specified parameter, which is difficult for the user to provide in practice. We estimate this parameter following [2] by taking the selected completely at random assumption and reformulate these two algorithms with this assumption. Furthermore, based on supervised algorithms Averaged One-Dependence Estimators (AODE), Hidden Naive Bayes (HNB) and Full Bayesian network Classifier (FBC), we extend these algorithms to PU learning algorithms (PAODE, PHNB and PFBC respectively). Experimental results on 20 UCI datasets show that the performance of the Bayesian algorithms for PU learning are comparable to corresponding supervised ones in most cases. Additionally, PNB and PFBC are more robust against unlabeled data, and PFBC generally performs the best.

Feature representation is essential to machine learning and text mining. In this paper, we present a feature coupling generalization (FCG) framework for generating new features from unlabeled data. It selects two special types of features, i.e., example-distinguishing features (EDFs) and class-distinguishing features (CDFs) from original feature set, and then generalizes EDFs into higher-level features based on their coupling degrees with CDFs in unlabeled data. The advantage is: EDFs with extreme sparsity in labeled data can be enriched by their co-occurrences with CDFs in unlabeled data so that the performance of these low-frequency features can be greatly boosted and new information from unlabeled can be incorporated. We apply this approach to three tasks in biomedical literature mining: gene named entity recognition (NER), protein-protein interaction extraction (PPIE), and text classification (TC) for gene ontology (GO) annotation. New features are generated from over 20 GB unlabeled PubMed abstracts. The experimental results on BioCreative 2, AIMED corpus, and TREC 2005 Genomics Track show that 1) FCG can utilize well the sparse features ignored by supervised learning. 2) It improves the performance of supervised baselines by 7.8 percent, 5.0 percent, and 5.8 percent, respectively, in the tree tasks. 3) Our methods achieve 89.1, 64.5 F-score, and 60.1 normalized utility on the three benchmark data sets.

Two birds with one stone: Classifying positive and unlabeled examples on uncertain data streams

A machine learning framework which uses unlabeled data from a related task domain in supervised classification tasks is described. The unlabeled data come from related domains, which share the same class labels or generative distribution as the labeled data. Patterns in the unlabeled data are learned via a neural network and transferred to the target domain from where the labeled data are generated, so as to improve the performance of the supervised learning task. We call this approach self-taught transfer learning from unlabeled data. We introduce a general-purpose feature learning algorithm producing features that retain information from the unlabeled data. Information preservation assures that the features obtained will be useful for improving the classification performance of the supervised tasks.

Person re-identification (re-ID) is an important topic in computer vision. We study the one-example re-ID task, where each identity has only one labeled example along with many unlabeled examples. In practice, for the unlabeled data, it is difficult to differentiate each person because of many conditions, such as low resolutions, occlusions, and lighting. In previous works, fine-grained information has been proven to be useful for supervised re-ID. To solve choosing a reliable, easy sample for self-paced learning, we exploit fine-grained features to metric the distances between labeled data and unlabeled data, the combination strategy of global and overlapping-part distance selecting more positive data for model training. In addition, the attention mechanism has been introduced to suppress the interruption of background. The training data are split into three parts, i.e., labeled data, pseudolabeled data, and instance-labeled data. First, the model is initialized by one-shot data for each identity. Then, pseudolabels are estimated from the unlabeled data and updating the model iteratively. The self-paced progressive sampling method is adopted to increase the number of the selected pseudolabeled candidates step by step. Notably, with pretrained model, on Market-1501, the rank-1 accuracy of our method is 86.0% which exceeds most other methods, experiment on two image-based datasets demonstrate promising results under one example re-ID setting.

Positive and unlabeled learning (PU learning) is a significant binary classification task in machine learning; it focuses on training accurate classifiers using positive data and unlabeled data. Most of the works in this area are based on a two-step strategy: the first step is to identify reliable negative examples from unlabeled examples, and the second step is to construct the classifiers based on the positive examples and the identified reliable negative examples using supervised learning methods. However, these methods always underutilize the remaining unlabeled data, which limits the performance of PU learning. Furthermore, many methods require the iterative solution of the formulated quadratic programming problems to obtain the final classifier, resulting in a large computational cost. In this paper, we propose a new method called the absolute value inequality support vector machine, which applies the concept of eccentricity to select reliable negative examples from unlabeled data and then constructs a classifier based on the positive examples, the selected negative examples, and the remaining unlabeled data. In addition, we apply a hyperparameter optimization technique to automatically search and select the optimal parameter values in the proposed algorithm. Numerical experimental results on ten real-world datasets demonstrate that our method is better than the other three benchmark algorithms.

With the exponential growth of biological data, labeling this kind of data becomes difficult and costly. Although unlabeled data are comparatively more plentiful than labeled ones, most supervised learning methods are not designed to use unlabeled data. Semi-supervised learning methods are motivated by the availability of large unlabeled datasets rather than a small amount of labeled examples. However, incorporating unlabeled data into learning does not guarantee an improvement in classification performance. This paper introduces an approach based on a model of semi-supervised learning, which is the self-training with a deep learning algorithm to predict missing classes from labeled and unlabeled data. In order to assess the performance of the proposed approach, two datasets are used with four performance measures: precision, recall, F-measure, and area under the ROC curve (AUC).

We present an algorithm for learning parametric classifiers on a partially labeled data manifold, based on a graph representation of the manifold. The unlabeled data are utilized by basing classifier learning on neighborhoods, formed via Markov random walks. The proposed algorithm yields superior performance on three benchmark data sets and the margin of improvements over existing semi-supervised algorithms is significant.

A growing specter in the rise of machine learning is whether the decisions made by machine learning models are fair. While research is already underway to formalize a machine-learning concept of fairness and to design frameworks for building fair models with sacrifice in accuracy, most are geared toward either supervised or unsupervised learning. Yet two observations inspired us to wonder whether semi-supervised learning might be useful to solve discrimination problems. First, previous study showed that increasing the size of the training set may lead to a better trade-off between fairness and accuracy. Second, the most powerful models today require an enormous of data to train which, in practical terms, is likely possible from a combination of labeled and unlabeled data. Hence, in this paper, we present a framework of fair semi-supervised learning in the pre-processing phase, including pseudo labeling to predict labels for unlabeled data, a re-sampling method to obtain multiple fair datasets and lastly, ensemble learning to improve accuracy and decrease discrimination. A theoretical decomposition analysis of bias, variance and noise highlights the different sources of discrimination and the impact they have on fairness in semi-supervised learning. A set of experiments on real-world and synthetic datasets show that our method is able to use unlabeled data to achieve a better trade-off between accuracy and discrimination.