3 papers
cs.CV2024
A Spitting Image: Modular Superpixel Tokenization in Vision Transformers
Marius Aasan, Odd Kolbjørnsen, Anne Schistad Solberg +1
Vision Transformer (ViT) architectures traditionally employ a grid-based approach to tokenization independent of the semantic content of an image. We propose a modular superpixel t…
cs.LG2023
Sequential Monte Carlo applied to virtual flow meter calibration
Anders T. Sandnes, Bjarne Grimstad, Odd Kolbjørnsen
Soft-sensors are gaining popularity due to their ability to provide estimates of key process variables with little intervention required on the asset and at a low cost. In oil and…
cs.LG2023
Multi-task neural networks by learned contextual inputs
Anders T. Sandnes, Bjarne Grimstad, Odd Kolbjørnsen
This paper explores learned-context neural networks. It is a multi-task learning architecture based on a fully shared neural network and an augmented input vector containing traina…