A multi-objective programming perspective to statistical learning problemsReport as inadecuate


A multi-objective programming perspective to statistical learning problems


A multi-objective programming perspective to statistical learning problems - Download this document for free, or read online. Document in PDF available to download.

It has been increasingly recognized that realistic problems often involve a tradeoff among many conflicting objectives. Traditional methods aimat satisfying multiple objectives by combining them into a global cost function, whichin most cases overlooks the underlying tradeoffs between the conflicting objectives.This raises the issue about how different objectives should be combined to yield afinal solution. Moreover, such approaches promise that the chosen overall objectivefunction is optimized over the training samples. However, there is no guarantee onthe performance in terms of the individual objectives since they are not consideredon an individual basis.Motivated by these shortcomings of traditional methods, the objective in thisdissertation is to investigate theory, algorithms, and applications for problems withcompeting objectives and to understand the behavior of the proposed algorithmsin light of some applications. We develop a multi-objective programming MOPframework for finding compromise solutions that are satisfactory for each of multiplecompeting performance criteria. The fundamental idea for our formulation, which werefer to as iterative constrained optimization ICO, evolves around improving oneobjective while allowing the rest to degrade. This is achieved by the optimization ofindividual objectives with proper constraints on the remaining competing objectives.The constraint bounds are adjusted based on the objective functions obtained inthe most recent iteration. An aggregated utility function is used to evaluate theacceptability of local changes in competing criteria, i.e., changes from one iterationto the next.Conflicting objectives arise in different contexts in many problems of speech andlanguage technologies. In this dissertation, we consider two applications. The firstapplication is language model LM adaptation, where a general LM is adapted to aspecific application domain so that the adapted LM is as close as possible to both thegeneral model and the application domain data. Language modeling and adaptation isused in many speech and language processing applications such as speech recognition,machine translation, part-of-speech tagging, parsing, and information retrieval.The second application is automatic language identification LID, where the standard detection performance evaluation measures false-rejection or miss and false-acceptance or false alarm rates for a number of languages are to be simultaneously minimized. LID systems might be used as a pre-processing stage for understandingsystems and for human listeners, and find applications in, for example, a hotel lobbyor an international airport where one might speak to a multi-lingual voice-controlledtravel information retrieval system.This dissertation is expected to provide new insights and techniques for accomplishing significant performance improvement over existing approaches in terms of the individual competing objectives. Meantime, the designer has a better control over what is achieved in terms of the individual objectives. Although many MOP approaches developed so far are formal and extensible to large number of competing objectives, their capabilities are examined only with two or three objectives. This is mainly because practical problems become significantly harder to manage when the number of objectives gets larger. We, however, illustrate the proposed framework with a larger number of objectives.



Georgia Tech Theses and Dissertations - School of Electrical and Computer Engineering Theses and Dissertations -



Author: Yaman, Sibel - -

Source: https://smartech.gatech.edu/







Related documents