- About
- Admissions
- Study at AUS
- Prospective Students
- Bachelor's Degrees
- Master's Degrees
- Doctoral Degrees
- Admission Publications
- International Students
- Contact Admissions
- Grants and Scholarships
- Sponsorship Liaison Services
- Testing Center
- New Student Guide
- File Completion
- New Student Orientation
- Payment Guide
- Executive Education
- Students with Disabilities
- Academics
- Life at AUS
- Research and Graduate Studies
- Contact Us
- Apply Now
- .
Lecture: Prediction of a Biological Response of Molecules from Their Chemical Properties (November 2012)
Dr. Dmitry Efimov is Visiting Assistant Professor in the AUS Department of Mathematics and Statistics.
Abstract:
Ensembles are often capable of greater prediction accuracy than any of their individual members. As a consequence of the diversity between individual base-learners, an ensemble will not suffer from over fitting. In this regard, development of the systematic and automatic approach for the evaluation of the solution is particularly important. Based on the mechanism of homogeneous ensembling (known also as bagging), we can construct a passport of the solution as unified validation trajectory against all available training data. Assuming that passports mimic closely the corresponding test solutions, we can use them for the consideration of many tasks including optimizations of blends and ensembles, calculation of the bias and any other tests as required. The reported results were obtained online during one of the most popular international data mining competition on the Kaggle platform in 2012.
For further details kindly contact [email protected].