One Computer Science MSc thesis to be presented on 30 October

Wednesday, 30 October there will be a master thesis presentation in Computer Science at Lund University, Faculty of Engineering.

The presentation will take place in E:4130 (LUCAS).

Note to potential opponents: Register as an opponent to the presentation of your choice by sending an email to the examiner for that presentation (firstname.lastname@cs.lth.se). Do not forget to specify the presentation you register for! Note that the number of opponents may be limited (often to two), so you might be forced to choose another presentation if you register too late. Registrations are individual, just as the oppositions are! More instructions are found on this page.

11:15-12:00 in E:4130 (LUCAS)

Presenter: Victor Schack
Title: Estimating performance of Autotrainers using Meta-Learning
Examiner: Jacek Malec
Supervisor: Marcus Klang (LTH)

Autotrainers are systems that automatically produce machine learning models for datasets. This work evaluates three autotrainers and records their performance over 208 datasets for six time limits. The evaluation covers aspects like, agreeableness, average performance, number of failed datasets and more. Up to 108 qualities for every dataset is used together with the performances to create a meta-dataset. The task is to use meta-learning to estimate the performance of autotrainers on new datasets. In doing this the details of meta-learning are examined, such as what qualities are important, how does importance differ for different time limits and autotrainers and how close to the actual performance the estimates are.

Link to popular science summary: To be uploaded