Daniel Klosa

Projekte

1. hyBit: Hydrogen for Bremen's industrial transformation (01.09.2022 - 28.02.2026)
2. DiSCO₂-Bremen: Datenbasierte und intelligente Simulation des Verkehrs zur CO₂-Reduktion in Bremen (01.07.2020 - 31.12.2022)

Abschlussarbeiten (Auswahl)vollständige Liste

1. Differenzierbare Architektursuche für die Vorhersage von Verkehrsdaten (Celina Groth)

Publikationen (Auswahl)vollständige Liste

1. D. Klosa, C. Büskens.
   Low Cost Evolutionary Neural Architecture Search (LENAS) Applied to Traffic Forecasting.
   21st IEEE International Conference on Machine Learning and Applications (ICMLA), 12.12.-14.12.2022, Nassau, Bahamas.
   Mach. Learn. Knowl. Extr., 5(3):830-846, 2023.

   DOI: 10.3390/make5030044

2. A. Mallek, D. Klosa, C. Büskens.
   Enhanced K-Nearest Neighbor Model For Multi-steps Traffic Flow Forecast in Urban Roads.
   8th IEEE International Smart Cities Conference, 26.09-29.09.2022.
   Paphos, Cyprus, IEEE Xplore, S. 1-5, 2022.

   DOI: 10.1109/ISC255366.2022.9921897

3. A. Mallek, D. Klosa, C. Büskens.
   Impact of Data Loss on Multi-Step Forecast of Traffic Flow in Urban Roads Using K-Nearest Neighbors.
   Sustainability, 14(18), 11232, 2022.

   DOI: https://doi.org/10.3390/su141811232

4. D. Klosa, A. Mallek, C. Büskens.
   Short-Term Traffic Flow Forecast Using Regression Analysis and Graph Convolutional Neural Networks.
   2021 IEEE 23rd Int Conf on High Performance Computing & Communications; 7th Int Conf on Data Science & Systems; 19th Int Conf on Smart City; 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys), 20.12-22.12.2021.
   Haikou, Hainan, China, IEEE Xplore, S. 1413-1418, 2021.

   DOI: 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00212