Szerző: Várhidi Renáta

Előadás címe: Synthetic foggy scene generation and evaluation pipeline Helyszín: Online Időpont: 2020.06.30. 17:10-17:30 Kivonat: We present an artificial intelligence based image processing pipeline for generating synthetic foggy scenes from sunny pictures by breaking down the problem into 1) sharp shadow removal, 2) depthmap generation, and 3) fog generation. Then …

Előadás címe: An adaptive method based on sequential tests to evaluate an image analyzing procedure Helyszín: Online Időpont: 2020.06.30. 15:35-15:55 Kivonat: Our aim is to decide whether an algorithm, that identifies certain diseases based on images, works with the desired efficiency. Based on sequential testing, we present an adaptive method …

Helyszín: Online, https://meet.google.com/pfp-yzpu-cbe Időpont: 2020.06.30. 15:55-16:15 Kivonat: We intend to investigate the changes of yearly mortalities in association with known epidemics.We are aimed at modeling these changes because those may affect the balance and provisioningof insurers. In the recent actuarial literature, the Lee-Carter [1] model and its variants are usedmost …

Előadás címe: Some results related to the GEHC project Helyszín: Online Időpont: 2020.06.30. 15:15-15:35 Kivonat: In the presentation, we first present the problems raised by GEHC and specify those the different groups have started to study. The University of Szeged has achieved results in minimizing observation cost by sequential methods. …

Előadás címe: COVID-19 modelling and control in a bank environment Helyszín: Online, https://meet.google.com/pfp-yzpu-cbe?authuser=0 Időpont: 2020.06.30. 16:50-17:10 Kivonat: The management of COVID-19 is a long term challenge, not only for the public health systems of countries, but also for other stakeholders, such as companies, especially for those who provide services to …

Előadás címe: Multivariate Markov-chain methods in monitoring health-care processes Helyszín: Online Időpont: 2020.06.30., 10:35 – 10:55 Kivonat: The application of Markov chain-based cost-optimal control charts on real-word data revealed that the method can be extended in areas where information is naturally multivariate (e.g. treatments, demographic variables etc.). From a mathematical …