Disputas: Håvard Kvamme

M.Sc. Håvard Kvamme ved Matematisk institutt vil forsvare sin avhandling for graden ph.d.

Time-to-Event Prediction with Neural Networks

Bildet av kandidaten.

Håvard Kvamme

Disputas

Universitetet i Oslo er for tiden stengt, og disputasen vil derfor bli strømmet direkte via Zoom. Verten vil moderere det digitale mens disputaslederen moderer disputasen. 
Ex auditorio-spørsmål: Disputasleder vil invitere til ex auditorio-spørsmål, og disse kan foretas enten skriftlig eller muntlig ved å klikke "Participants -> Raise hand". 

Delta på disputasen/Join the PhD defence (deactivated)
Møtet åpnes for deltakelse like før 11.00, og stenges for nye deltagere ca. 15 minutter etter at disputasen har begynt.

Download zoom

Digital utgave av avhandlingen

Send inn forespørsel for å få tilgang til avhandlingen. / Submit the request to get access to the thesis.

Thesis request

Prøveforelesning

Prøveforelesningstittel: "Model-based vs. black box learning".

Digitalt opptak av prøveforelesning

Bedømmelseskomité

  • Professor Harald Binder, Albert-Ludwigs-Universität Freiburg

  • Professor Jan Terje Kvaløy, Universitetet i Stavanger

  • Professor Geir Storvik, Universitet i Oslo

Leder av disputas

Professor Erlend Wold, Matematisk institutt, Universitet i Oslo

Veiledere

  • Professor Ørnulf Borgan, Universitetet i Oslo

  • Førsteamanuensis Ida Scheel, Universitetet i Oslo

  • Assisterende forskningssjef Kjersti Aas, Norsk Regnesentral 

Sammendrag

In the last decades the analytical value of data has really become apparent and the amount of data collected has vastly increased. This enables us to approach problems in more data driven manners. In the thesis, I have combined recent developments in machine learning with statistical methods to better answer the question: “When in the future will a given event occur?”

The first part of the thesis was done in collaboration with the Norwegian bank DNB. We created new methods for predicting when in the future customers will default on their mortgage loans. By investigating the historical balances of the customers’ checking accounts, savings accounts and credit cards, we found that we could improve on existing methods for predicting mortgage defaults.

In the second part of the thesis, our attention was directed toward more general methodology that may be applied to a number of problems. Our proposed improvements were illustrated using a selection of available datasets, ranging from how gene and protein expression profiles affect the mortality of breast cancer patients, to how customer information can help determine if customers are likely to continue to subscribe to a music streaming service.

For mer informasjon

Kontakt Matematisk institutt

Publisert 14. mai 2020 14:36 - Sist endret 28. mai 2020 11:31