Robustly and Optimally Controlled Training Of neural Networks II (OCTON II)
Forskningsprojekt, 2020 – 2025

This project aims at focusing on the development novel methods that accounts for non-traditional training objectives (other than mean square prediction error) and corrupted data sequence. This project is expected to result in faster and more accurate training solutions (classification, parameter estimation, short time prediction, tracking) than the currently available ones. The methods developed are application free and concentrates on the triplet of interpretability, robustness and network optimization via deeplearners (DNN).

Deltagare

Vincent Szolnoky (kontakt)

Chalmers, Matematiska vetenskaper, Tillämpad matematik och statistik

Rebecka Jörnsten

Chalmers, Matematiska vetenskaper, Tillämpad matematik och statistik

Balázs Adam Kulcsár

Chalmers, Elektroteknik, System- och reglerteknik

Finansiering

Centiro

Finansierar Chalmers deltagande under 2020–2025

Relaterade styrkeområden och infrastruktur

Transport

Styrkeområden

Publikationer

2024

Controlled Descent Training

Artikel i vetenskaplig tidskrift
2023

Deep Q-learning: a robust control approach

Artikel i vetenskaplig tidskrift

Mer information

Senast uppdaterat

2022-05-03