We discuss a new framework for solving extreme classification (i.e., learning problems with an extremely large label space), in which we reduce the original problem to a structured prediction problem. Thanks to this we can obtain learning algorithms that work under a strict time and space budget. We mainly focus on a recently introduced algorithm, referred to as LTLS, which is to our best knowledge the first truly logarithmic time and space (in the number of labels) method for extreme classification. We compare this algorithm with two other approaches that also rely on transformation to structured prediction problems. The first algorithm encodes original labels as binary sequences. The second algorithm follows the label tree approach. The comparison shows the trade-off between computational complexity (in time and space) and predictive performance.

There is a strong research eort towards developing models that can achieve state-of-the-art results without sacrificing interpretability and simplicity. One of such is recently proposed Recursive Random Support Vector Machine (R2SVM) model, which is composed of stacked linear models. R2SVM was reported to learn deep representations outperforming many strong classiffiers like Deep Convolutional Neural Network. In this paper we try to analyze it both from theoretical and empirical perspective and show its important limitations.Analysis of similar model Deep Representation Extreme Learning Machine (DrELM) is also included. It is concluded that models in its current form achieves lower accuracy scores than Support Vector Machine with Radial Basis Function kernel.

Deep neural networks are currently among the most commonly used classifiers. Despite easily achieving very good performance, one of the best selling points of these models is their modular design – one can conveniently adapt their architecture to specific needs, change connectivity patterns, attach specialised layers, experiment with a large amount of activation functions, normalisation schemes and many others. While one can find impressively wide spread of various configurations of almost every aspect of the deep nets, one element is, in authors’ opinion, underrepresented – while solving classification problems, vast majority of papers and applications simply use log loss. In this paper we try to investigate how particular choices of loss functions affect deep models and their learning dynamics, as well as resulting classifiers robustness to various effects. We perform experiments on classical datasets, as well as provide some additional, theoretical insights into the problem. In particular we show that L1 and L2 losses are, quite surprisingly, justified classification objectives for deep nets, by providing probabilistic interpretation in terms of expected misclassification. We also introduce two losses which are not typically used as deep nets objectives and show that they are viable alternatives to the existing ones.

Object ranking is one of the most relevant problems in the realm of preference learning and ranking. It is mostly tackled by means of two different techniques, often referred to as pairwise and pointwise ranking. In this paper, we present a case study in which we systematically compare two representatives of these techniques, a method based on the reduction of ranking to binary classification and so-called expected rank regression (ERR). Our experiments are meant to complement existing studies in this field, especially previous evaluations of ERR. And indeed, our results are not fully in agreement with previous findings and partly support different conclusions.

In this work the subject of the application of clustering as a knowledge extraction method from real-world data is discussed. The authors analyze an influence of different clustering parameters on the quality of the created structure of rules clusters and the efficiency of the knowledge mining process for rules / rules clusters. The goal of the experiments was to measure the impact of clustering parameters on the efficiency of the knowledge mining process in rulebased knowledge bases denoted by the size of the created clusters or the size of the representatives. Some parameters guarantee to produce shorter/longer representatives of the created rules clusters as well as smaller/greater clusters sizes.

Continuous vector representations, as a distributed representations for words have gained a lot of attention in Natural Language Processing (NLP) field. Although they are considered as valuable methods to model both semantic and syntactic features, they still may be improved. For instance, the open issue seems to be to develop different strategies to introduce the knowledge about the morphology of words. It is a core point in case of either dense languages where many rare words appear and texts which have numerous metaphors or similies. In this paper, we extend a recent approach to represent word information. The underlying idea of our technique is to present a word in form of a bag of syllable and letter n-grams. More specifically, we provide a vector representation for each extracted syllable-based and letter-based n-gram, and perform concatenation. Moreover, in contrast to the previous method, we accept n-grams of varied length n. Further various experiments, like tasks-word similarity ranking or sentiment analysis report our method is competitive with respect to other state-of-theart techniques and takes a step toward more informative word representation construction.

Robust mixture models approaches, which use non-normal distributions have recently been upgraded to accommodate data with fixed bounds. In this article we propose a new method based on uniform distributions and Cross-Entropy Clustering (CEC). We combine a simple density model with a clustering method which allows to treat groups separately and estimate parameters in each cluster individually. Consequently, we introduce an effective clustering algorithm which deals with non-normal data.

Recent methods for learning word embeddings, like GloVe or Word2-Vec, succeeded in spatial representation of semantic and syntactic relations. We extend GloVe by introducing separate vectors for base form and grammatical form of a word, using morphosyntactic dictionary for this. This allows vectors to capture properties of words better. We also present model results for word analogy test and introduce a new test based on WordNet.

Due to rapid growth of computational power and demand for faster and more optimal solution in today's manufacturing, machine learning has lately caught a lot of attention. Thanks to it's ability to adapt to changing conditions in dynamic environments it is perfect choice for processes where rules cannot be explicitly given. In this paper proposes on-line supervised learning approach for optimal scheduling in manufacturing. Although supervised learning is generally not recommended for dynamic problems we try to defeat this conviction and prove it's viable option for this class of problems. Implemented in multi-agent system algorithm is tested against multi-stage, multi-product flow-shop problem. More specifically we start from dening considered problem. Next we move to presentation of proposed solution. Later on we show results from conducted experiments and compare our approach to centralized reinforcement learning to measure algorithm performance.

Mutation testing is considered as one of the most effective quality improvement technique by assessing the strength of the actual test suite. If no test is able to kill a given mutant, this means that the tests are not strong enough and we need to write additional one that will be able to kill this mutant. However, mutation testing is very time consuming. In this paper we investigate if it is possible to reduce the scope of the mutation analysis by running it only on the new or changed part of the code. Using data from the real open-source projects we analyze if there is a relation between mutation scope reduction and effectiveness of the mutation analysis.