Abstract Modern real-world control problems call for continuous control domains and robust, sample efficient and explainable control frameworks. We are presenting a framework for recursively composing control skills to solve compositional and progressively complex tasks. The framework promotes reuse of skills, and as a result quick adaptability to new tasks. The decision tree can be observed, providing insight into the agents’ behavior. Furthermore, the skills can be transferred, modified or trained independently, which can simplify reward shaping and increase training speeds considerably. This paper is concerned with efficient composition of control algorithms using reinforcement learning and soft attention. Compositional and temporal abstraction is the key to improving learning and planning in reinforcement learning. Our Thompson sampling inspired soft-attention model is demonstrated to efficiently solve the composition problem.

We, as humans, are constantly in relation with our environment. Sudden changes in our living media may alter the way we perceive ourselves and our environment in various ways. Coronavirus (COVID-19) outbreak is a great example of a sudden change. The outbreak influenced means of work, study, socialization, and communication in unprecedented ways. In our study, we investigate the topic dynamics of Twitter content sharing for the Republic of Turkey. We have analyzed 1.3 million tweets containing the keyword “korona” shared between February 24, 2020 and May 2, 2020. Our study has three key outcomes. The first one is, after the official announcement of first COVID-19 case in Turkey, rate of COVID-19-related content sharing decreases and hygiene-, lifestyle-, and anxiety-related tweets increase. Second, we see that a number of friends and followers influence content sharing dynamics where accounts sharing COVID-19 News-related content have more followers compared to accounts that share the remaining contents. Finally, motivated by the World Bank’s Poverty Monitoring Technical Note, we inquired the effect of income on content sharing and found that GDP per capita of the author’s city is more influential on COVID-19 News-related content sharing compared to the population and COVID-19 cases per 1 00 000 people. The lower the GDP per capita, the higher the COVID-19 News-related content sharing is. Also, our model indicates that lower income and population along with high rates of COVID-19 cases per 1 00 000 people are associated with increased COVID-19 News-related content sharing.

Over the last decade, Distributed Denial of Service (DDoS) attacks have been employed to cause huge financial and prestige loss to different kinds of e-business. Attackers also target governmental websites using DDoS attacks as a new weapon in the world of cyber war. The importance of the issue has inspired many researchers from academia and the industry to provide solutions to this type of challenging attack. In this study, we simulated DDoS attacks in a virtual lab and then collected firewall logs from the Security Information and Event Management (SIEM) platform of a company in the field of security management solutions. We extracted 14 research features from firewall logs and applied a SMO algorithm to train our data using 10 fold cross-validation. The SMO with PolyKernel was able to create a prediction model without any false alarm. We also tested our model with two different datasets. This research is an ongoing multistep study. Future research will concentrate on online DDoS detection.

In this paper, we propose a nonconvex online Support Vector Machine (SVM) algorithm (LASVM-NC) based on the Ramp Loss, which has the strong ability of suppressing the influence of outliers. Then, again in the online learning setting, we propose an outlier filtering mechanism (LASVM-I) based on approximating nonconvex behavior in convex optimization. These two algorithms are built upon another novel SVM algorithm (LASVM-G) that is capable of generating accurate intermediate models in its iterative steps by leveraging the duality gap. We present experimental results that demonstrate the merit of our frameworks in achieving significant robustness to outliers in noisy data classification where mislabeled training instances are in abundance. Experimental evaluation shows that the proposed approaches yield a more scalable online SVM algorithm with sparser models and less computational running time, both in the training and recognition phases, without sacrificing generalization performance.We also point out the relation between nonconvex optimization and min-margin active learning.

This paper is concerned with the class imbalance problem which has been known to hinder the learning performance of classification algorithms. The problem occurs when there are significantly less number of observations of the target concept. Various real-world classification tasks, such as medical diagnosis, text categorization and fraud detection suffer from this phenomenon. The standard machine learning algorithms yield better prediction performance with balanced datasets. In this paper, we demonstrate that active learning is capable of solving the class imbalance problem by providing the learner more balanced classes. We also propose an efficient way of selecting informative instances from a smaller pool of samples for active learning which does not necessitate a search through the entire dataset. The proposed method yields an efficient querying system and allows active learning to be applied to very large datasets. Our experimental results show that with an early stopping criteria, active learning achieves a fast solution with competitive prediction performance in imbalanced data classification.

We propose a novel multiclass classification algorithm Gentle Adaptive Multiclass Boosting Learning (GAMBLE). The algorithm naturally extends the two class Gentle AdaBoost algorithm to multiclass classification by using the multiclass exponential loss and the multiclass response encoding scheme. Unlike other multiclass algorithms which reduce the K-class classification task to K binary classifications, GAMBLE handles the task directly and symmetrically, with only one committee classifier. We formally derive the GAMBLE algorithm with the quasi-Newton method, and prove the structural equivalence of the two regression trees in each boosting step. To scale up to large datasets, we utilize the generalized Query By Committee (QBC) active learning framework to focus learning on the most informative samples. Our empirical results show that with QBC-style active sample selection, we can achieve faster training time and potentially higher classification accuracy. GAMBLE's numerical superiority, structural elegance and low computation complexity make it highly competitive with state-of-the-art multiclass classification algorithms.

The class imbalance problem has been known to hinder the learning performance of classification algorithms. Various real-world classification tasks such as text categorization suffer from this phenomenon. We demonstrate that active learning is capable of solving the problem.

Identification of distinct clusters of documents in text collections has traditionally been addressed by making the assumption that the data instances can only be represented by homogeneous and uniform features. Many real-world data, on the other hand, comprise of multiple types of heterogeneous interrelated components, such as web pages and hyperlinks, online scientific publications and authors and publication venues to name a few. In this paper, we present KSVMeans, a clustering algorithm for multi-type interrelated datasets that integrates the well known K-Means clustering with the highly popular Support Vector Machines. The experimental results on authorship analysis of two real world web-based datasets show that K-SVMeans can successfully discover topical clusters of documents and achieve better clustering solutions than homogeneous data clustering.

Traditional clustering algorithms work on "flat" data, making the assumption that the data instances can only be represented by a set of homogeneous and uniform features. Many real world data, however, is heterogeneous in nature, comprising of multiple types of interrelated components. We present a clustering algorithm, K-SVMeans, that integrates the well known K-Means clustering with the highly popular Support Vector Machines(SVM) in order to utilize the richness of data. Our experimental results on authorship analysis of scientific publications show that K-SVMeans achieves better clustering performance than homogeneous data clustering.

Very high dimensional learning systems become theoretically possible when training examples are abundant. The computing cost then becomes the limiting factor. Any efficient learning algorithm should at least take a brief look at each example. But should all examples be given equal attention? This contribution proposes an empirical answer. We first present an online SVM algorithm based on this premise. LASVM yields competitive misclassification rates after a single pass over the training examples, outspeeding state-of-the-art SVM solvers. Then we show how active example selection can yield faster training, higher accuracies, and simpler models, using only a fraction of the training example labels.