Statistical learning theory

About: Statistical learning theory is a research topic. Over the lifetime, 1618 publications have been published within this topic receiving 158033 citations.

One Size Does Not Fit All: The Limits of Structure-Based Models in Drug Discovery

Abstract: A major goal in computational chemistry has been to discover the set of rules that can accurately predict the binding affinity of any protein-drug complex, using only a single snapshot of its three-dimensional structure. Despite the continual development of structure-based models, predictive accuracy remains low, and the fundamental factors that inhibit the inference of all-encompassing rules have yet to be fully explored. Using statistical learning theory and information theory, here we prove that even the very best generalized structure-based model is inherently limited in its accuracy, and protein-specific models are always likely to be better. Our results refute the prevailing assumption that large data sets and advanced machine learning techniques will yield accurate, universally applicable models. We anticipate that the results will aid the development of more robust virtual screening strategies and scoring function error estimations.

Online learning via sequential complexities

Abstract: We consider the problem of sequential prediction and provide tools to study the minimax value of the associated game. Classical statistical learning theory provides several useful complexity measures to study learning with i.i.d. data. Our proposed sequential complexities can be seen as extensions of these measures to the sequential setting. The developed theory is shown to yield precise learning guarantees for the problem of sequential prediction. In particular, we show necessary and sufficient conditions for online learnability in the setting of supervised learning. Several examples show the utility of our framework: we can establish learnability without having to exhibit an explicit online learning algorithm.

Using support vector machine for materials design

Abstract: Materials design is the most important and fundamental work on the background of materials genome initiative for global competitiveness proposed by the National Science and Technology Council of America. As far as the methodologies of materials design, besides the thermodynamic and kinetic methods combing databases, both deductive approaches so-called the first principle methods and inductive approaches based on data mining methods are gaining great progress because of their successful applications in materials design. In this paper, support vector machine (SVM), including support vector classification (SVC) and support vector regression (SVR) based on the statistical learning theory (SLT) proposed by Vapnik, is introduced as a relatively new data mining method to meet the different tasks of materials design in our lab. The advantage of using SVM for materials design is discussed based on the applications in the formability of perovskite or BaNiO3 structure, the prediction of energy gaps of binary compounds, the prediction of sintered cold modulus of sialon-corundum castable, the optimization of electric resistances of VPTC semiconductors and the thickness control of In2O3 semiconductor film preparation. The results presented indicate that SVM is an effective modeling tool for the small sizes of sample sets with great potential applications in materials design.

Sphere-structured support vector machines for multi-class pattern recognition

Abstract: Support vector machines (SVM) are learning algorithms derived from statistical learning theory. The SVM approach was originally developed for binary classification problems. For solving multi-class classification problem, there are some methods such as one-against-rest, one-against-one, all-together and so on. But the computing time of all these methods are too long to solve large scale problem. In this paper SVMs architectures for multi-class problems are discussed, in particular we provide a new algorithm called sphere-structured SVMs to solve the multi-class problem. We show the algorithm in detail and analyze its characteristics. Not only the number of convex quadratic programming problems in sphere-structured SVMs is small, but also the number of variables in each programming is least. The computing time of classification is reduced. Otherwise, the characteristics of sphere-structured SVMs make expand data easily.

Research on bearing life prediction based on support vector machine and its application

Abstract: Life prediction of rolling element bearing is the urgent demand in engineering practice, and the effective life prediction technique is beneficial to predictive maintenance. Support vector machine (SVM) is a novel machine learning method based on statistical learning theory, and is of advantage in prediction. This paper develops SVM-based model for bearing life prediction. The inputs of the model are features of bearing vibration signal and the output is the bearing running time-bearing failure time ratio. The model is built base on a few failed bearing data, and it can fuse information of the predicted bearing. So it is of advantage to bearing life prediction in practice. The model is applied to life prediction of a bearing, and the result shows the proposed model is of high precision.