Batch Mode Active Sampling Based on Marginal Probability Distribution Matching

Support vector machine (SVM) active learning is one popular and successful technique for relevance feedback in content-based image retrieval (CBIR). Despite the success, conventional SVM active learning has two main drawbacks. First, the performance of SVM is usually limited by the number of labeled examples. It often suffers a poor performance for the small-sized labeled examples, which is the case in relevance feedback. Second, conventional approaches do not take into account the redundancy among examples, and could select multiple examples that are similar (or even identical). In this work, we propose a novel scheme for explicitly addressing the drawbacks. It first learns a kernel function from a mixture of labeled and unlabeled data, and therefore alleviates the problem of small-sized training data. The kernel will then be used for a batch mode active learning method to identify the most informative and diverse examples via a min-max framework. Two novel algorithms are proposed to solve the related combinatorial optimization: the first approach approximates the problem into a quadratic program, and the second solves the combinatorial optimization approximately by a greedy algorithm that exploits the merits of submodular functions. Extensive experiments with image retrieval using both natural photo images and medical images show that the proposed algorithms are significantly more effective than the state-of-the-art approaches. A demo is available at http://msm.cais.ntu.edu.sg/LSCBIR/.

Active learning for networked data that focuses on predicting the labels of other nodes accurately by knowing the labels of a small subset of nodes is attracting more and more researchers because it is very useful especially in cases, where labeled data are expensive to obtain. However, most existing research either only apply to networks with assortative community structure or focus on node attribute data with links or are designed for working in single mode that will work at a higher learning and query cost than batch active learning in general. In view of this, in this paper, we propose a batch mode active learning method which uses information-theoretic techniques and random walk to select which nodes to label. The proposed method requires only network topology as its input, does not need to know the number of blocks in advance, and makes no initial assumptions about how the blocks connect. We test our method on two different types of networks: assortative structure and diassortative structure, and then compare our method with a single mode active learning method that is similar to our method except for working in single mode and several simple batch mode active learning methods using information-theoretic techniques and simple heuristics, such as employing degree or betweenness centrality. The experimental results show that the proposed method in this paper significantly outperforms them.

Deep learning has emerged as a powerful tool in various domains, revolutionising machine learning research. However, one persistent challenge is the scarcity of labelled training data, which hampers the performance and generalisation of deep learning models. To address this limitation, researchers have developed innovative methods to overcome data scarcity and enhance deep model learning capabilities. Two prevalent techniques that have gained significant attention are transfer learning and self-supervised learning. Transfer learning leverages knowledge learned from pre-training on a large-scale dataset, such as ImageNet, and applies it to a target task with limited labelled data. This approach allows models to benefit from the learned representations and effectively transfer knowledge to new tasks, resulting in improved learning performance and generalisation. On the other hand, self-supervised learning focuses on training models using pretext tasks that do not require manual annotation, allowing them to learn valuable representations from large amounts of unlabelled data. These learned representations can then be fine-tuned for downstream tasks, mitigating the need for extensive labelled data. In recent years, transfer and self-supervised learning have found applications in various fields, including medical image processing, video recognition, and natural language processing. These approaches have demonstrated remarkable achievements, enabling breakthroughs in areas such as disease diagnosis, object recognition, and language understanding. However, while these methods offer numerous advantages, they also have limitations. For example, transfer learning may face domain mismatch issues between the pre-training and target domains, while self-supervised learning requires careful design of pretext tasks to ensure meaningful representations. This review paper explores the recent applications of these pre-training methods in various fields within the past three years. It delves into the advantages and limitations of each approach, assesses the performance of models employing these techniques, and identifies potential directions for future research. By providing a comprehensive review of current pre-training methods, this article offers guidance for selecting the best technique for specific deep learning applications to address the data scarcity issue.

We organized a data mining challenge on “active learning” for IJCNN/WCCI 2010, addressing machine learning problems where labeling data is expensive, but large amounts of unlabeled data are available at low cost. Examples include handwriting and speech recognition, document classification, vision tasks, drug design using recombinant molecules and protein engineering. Such problems might be tackled from different angles: learning from unlabeled data or active learning. In the former case, the algorithms must satisfy themselves with the limited amount of labeled data and capitalize on the unlabeled data with semi-supervised learning methods. Several challenges have addressed this problem in the past. In the latter case, the algorithms may place a limited number of queries to get new sample labels. The goal in that case is to optimize the queries and the problem is referred to as active learning. While the problem of active learning is of great importance, organizing a challenge in that area is non trivial. This is the problem we have addressed, and we describe our approach in this paper. The “active learning” challenge is part of the WCCI 2010 competition program (http://www.wcci2010. org/competition-program). The website of the challenge remains open for submission of new methods beyond the termination of the challenge as a resource for students and researchers (http://clopinet.com/al).

Linear regression methods are commonly used by both researchers and data scientists due to their interpretability and their reduced likelihood of overfitting. However, these methods can still perform poorly if little labeled training data is available. Typical methods used to overcome a lack of labeled training data somehow involve exploiting an outside source of labeled data or large amounts of unlabeled data. This includes areas such as active learning, semi-supervised learning and transfer learning, but in many domains these approaches are not always applicable because they require either a mechanism to label data, large amounts of unlabeled data or additional sources of sufficiently related data. In this paper we explore an alternative, non-data centric approach. We allow the user to guide the learning system through three forms of feature-level guidance which constrain the parameters of the regression function. Such guidance is unlikely to be perfectly accurate, so we derive methods which are robust to some amounts of noise, a property we formally prove for one of our methods.

Intrusion alerts continue to grow in volume, variety, and complexity. Its cryptic nature requires substantial time and expertise to interpret the intended consequence of observed malicious actions. To assist security analysts in effectively diagnosing what alerts mean, this work develops a novel machine learning approach that translates alert descriptions to intuitively interpretable Action-Intent-Stages (AIS) with only 1% labeled data. We combine transfer learning, active learning, and pseudo labels and develop the Pseudo-Active Transfer Learning (PATRL) process. The PATRL process begins with an unsupervised-trained language model using MITRE ATT&CK, CVE, and IDS alert descriptions. The language model feeds to an LSTM classifier to train with 1% labeled data and is further enhanced with active learning using pseudo labels predicted by the iteratively improved models. Our results suggest PATRL can predict correctly for 85% (top-1 label) and 99% (top-3 labels) of the remaining 99% unknown data. Recognizing the need to build confidence for the analysts to use the model, the system provides Monte-Carlo Dropout Uncertainty and Pseudo-Label Convergence Score for each of the predicted alerts. These metrics give the analyst insights to determine whether to directly trust the top-1 or top-3 predictions and whether additional pseudo labels are needed. Our approach overcomes a rarely tackled research problem where minimal amounts of labeled data do not reflect the truly unlabeled data's characteristics. Combining the advantages of transfer learning, active learning, and pseudo labels, the PATRL process translates the complex intrusion alert description for the analysts with confidence.

This paper studies the relationship prediction problem in multi-network scenarios, aiming to overcome the network sparsity challenge where the labeled data (connected node pairs) are much less than the unlabeled data (unconnected node pairs). The TAQIL framework is proposed by using transfer learning to get knowledge from the related source networks and then use active learning to query the labels of the most informative instances from the oracle in the target network. A new query function is also proposed in order to better use the parameters output by the transfer learning method. The alternate use of transfer learning and active learning allows adaptive transfer of knowledge across multiple networks to mitigate cold start and meantime improve the prediction accuracy with active queries in the target network. The experimental results on both non-network datasets and network datasets demonstrate the significant improvement in prediction accuracy compared with several benchmark methods and related state-of-art methods.

To assess the use of deep learning (DL) for computer-assisted glaucoma identification, and the impact of training using images selected by an active learning strategy, which minimizes labelling cost. Additionally, this study focuses on the explainability of the glaucoma classifier. This original investigation pooled 8433 retrospectively collected and anonymized colour optic disc-centred fundus images, in order to develop a deep learning-based classifier for glaucoma diagnosis. The labels of the various deep learning models were compared with the clinical assessment by glaucoma experts. Data were analysed between March and October 2018. Sensitivity, specificity, area under the receiver operating characteristic curve (AUC), and amount of data used for discriminating between glaucomatous and non-glaucomatous fundus images, on both image and patient level. Trained using 2072 colour fundus images, representing 42% of the original training data, the trained DL model achieved an AUC of 0.995, sensitivity and specificity of, respectively, 98.0% (CI 95.5%-99.4%) and 91% (CI 84.0%-96.0%), for glaucoma versus non-glaucoma patient referral. These results demonstrate the benefits of deep learning for automated glaucoma detection based on optic disc-centred fundus images. The combined use of transfer and active learning in the medical community can optimize performance of DL models, while minimizing the labelling cost of domain-specific mavens. Glaucoma experts are able to make use of heat maps generated by the deep learning classifier to assess its decision, which seems to be related to inferior and superior neuroretinal rim (within ONH), and RNFL in superotemporal and inferotemporal zones (outside ONH).

Developing intelligent learning algorithms under the constraint of limited manual labor is a fundamental research challenge and a problem of immense practical importance. Active learning algorithms alleviate this problem by automatically identifying the salient and exemplar instances from large amounts of unlabeled data; this tremendously reduces the human annotation effort as only a small subset of the samples, identi ed by the algorithm, needs to be labeled manually. Further, the unprecedented increase in the volume of digital data has necessitated the usage of multiple, independent computers for its storage and processing, in a given application. The need of the hour is therefore an active learning framework which can operate in a distributed setup, where the unlabeled data is partitioned across multiple computers. In this paper, we propose a novel algorithm to address this important challenge. Our algorithm requires minimal communication among the computers (over which the data is stored) and also enjoys nice theoretical properties. Our extensive empirical studies on a variety of challenging, real-world vision datasets, from different application domains, corroborate the potential of the proposed framework.

In this paper, we study the edge artificial intelligence (AI) techniques for industry 4.0. More specifically, we assume fog computing takes place on the edge of Industrial Internet of Things (IIoT) networks. We provide details about the three main edge AI techniques that can contribute to the future industrial applications. In particular, we deal with the active learning (AL), transfer learning (TL) and federated learning (FL), where AL is used to deal with the problem of unlabeled data, the TL is used to start training with a pre-trained model and the FL is a distributed solution to provide privacy. Finally, their combination is developed too that we name it federated active transfer learning (FATL). Simulation results are carried out that reveal the gain of each solution and their FATL combination. The deployment of FATL in IIoT networking standards such as IEEE P2805 is described too that can be extended as our future work.

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

Dynamic multiobjective optimization problems (DMOPs) aim to optimize multiple (often conflicting) objectives that are changing over time. Recently, there are a number of promising algorithms proposed based on transfer learning methods to solve DMOPs. However, it is very challenging to reduce the negative effect in transfer learning and find more effective transferred solutions. To fill this research gap, this article proposes a clustering-based transfer (CBT) learning method to solve DMOPs. When the environment changes, two novel operations (clustering-based selection (CBS) and CBT) are used to guide knowledge transfer. Specifically, CBS aims to find a population with nondominated solutions and dominated solutions as the training data for the new environment. Then, CBT further collects the previous Pareto-optimal solutions and some noise solutions as the training data for the previous environment. Two training data sets from different environments are, respectively, divided into multiple clusters and transfer learning is conducted on two similar clusters with high probability to reduce the negative effect, which can train an accurate prediction model to identify the promising solutions for the new environment. Empirical studies have been conducted on 14 benchmark DMOPs and one real-life path planning problem of unmanned air/ground vehicles, which validate the effectiveness of our proposed method. Especially, our method can significantly reduce negative transfer on 12 out of 14 cases when compared with direct transfer learning.

In many human activity recognition systems the size of the unlabeled training data may be significantly large due to expensive human effort required for data annotation. Moreover, the insufficient data collection process from heterogenous sources may cause dissimilarities between training and testing data. To address these limitations, a novel probabilistic approach that combines learning using privileged information (LUPI) and active learning is proposed. A pool-based privileged active learning approach is presented for semi-supervising learning of human activities from multimodal labeled and unlabeled data. Both uncertainty and distance from the decision boundary are used as query inference strategies for selecting an unlabeled observation and querying its label. Experimental results in four publicly available datasets demonstrate that the proposed method can identify complex human activities with high accuracy.

We propose a novel unsupervised transfer learning framework that utilises unlabelled auxiliary data to quantify and select the most relevant transferrable knowledge for recognising a target object class from the background given very limited training target samples. Unlike existing transfer learning techniques, our method does not assume that auxiliary data are labelled, nor the relationships between target and auxiliary classes are known a priori. Our unsupervised transfer learning is formulated by a novel kernel adaptation transfer (KAT) learning framework, which aims to (a) extract general knowledge about how more structured objects are visually distinctive from cluttered background regardless object class, and (b) more importantly, perform selective transfer of knowledge extracted from the auxiliary data to minimise negative knowledge transfer suffered by existing methods. The effectiveness and efficiency of the proposed approach is demonstrated by performing one-class object recognition (object vs. background) task using the Caltech256 dataset.

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.