Deep Learning Research

PhD Student in Machine Learning
Interests:
Neural Operators | Remote Sensing | Climate

Developing novel deep learning methodology and applying it to remote sensing and climate problems as a PhD Fellow at UiT The Arctic University of Norway.

Explore Research View Portfolio
Neural Network

About Me

PhD Student & AI Researcher

I'm a PhD student at UiT The Arctic University of Norway in Tromsø, working at the intersection of artificial intelligence and Earth system science. My research focuses on the enhancement of neural operator methodology and applying machine learning to sea ice forecasting, climate modeling, and Earth observation.

I contribute to several international research projects including IceNet (AI-powered sea ice forecasting), Visual Intelligence, and KnowEarth, developing novel AI approaches for understanding and utilizing state-of-the-art deep learning methods in the context of climate research and beyond.

Within my research at the Arctic University of Norway, I closely collaborate with other leading research institutes, including:
  - The Norwegian Computing Center
  - University of Southern Brittany
  - University of Cambridge
  - The Alan Turing Institute
  - The British Antarctic Survey
  - The MET Office

Core Expertise

Physics Mathematics Machine Learning Deep Learning Neural Operators Python PyTorch High-Performance Computing Climate Science Remote Sensing
Lars Uebbing

Research Focus

🧊 Sea Ice Forecasting

Developing AI-powered forecasting systems for Arctic sea ice, contributing to IceNet and advancing seasonal to daily sea ice predictions using deep learning.

🛰️ Earth Observation

Applying machine learning to satellite imagery and Earth observation data for environmental monitoring, climate research, and Arctic studies.

🌍 Climate Science & AI

Integrating artificial intelligence with climate science to understand environmental changes, particularly in polar regions and global climate systems.

Publications

Main Publications

NOFE - Neural Operator Function Embedding

Lars Uebbing, et al.

Proceeding of AAAI, 2026

A novel approach for functional dimensionality reduction leveraging core properties of neural operators to provide continuous and consistent embeddings.

Paper Code Supplementary

Investigating the Impact of Feature Reduction for Deep Learning-based Seasonal Sea Ice Forecasting

Lars Uebbing, et al.

Proceedings of Machine Learning Research (PMLR), 2025

Explainability study of IceNet, investigating feature importance for deep learning based forecasting of sea ice extreme events.

Paper Code

Co-Authorships

Placeholder

Primary Author, Lars Uebbing, et al.

Environmental Data Science, 2024

probably something hat harald publishes.

Paper Data

Research Projects

IceNet

Contributing to a deep learning sea ice forecasting system developed by British Antarctic Survey and The Alan Turing Institute. IceNet provides 6-month seasonal forecasts and daily operational predictions for Arctic sea ice.

Deep Learning Sea Ice Forecasting Python
Project Website GitHub

Visual Intelligence

Norwegian centre for research-based innovation focusing on next-generation deep learning methodology for visual data across medicine, marine science, energy, and Earth observation.

Computer Vision Deep Learning Earth Observation
Project Website

Machine Learning and Human Knowledge

Machine learning and human knowledge project focusing on integrating AI techniques with domain expertise for Earth system understanding and environmental research.

Machine Learning Earth Science Knowledge Integration
Project Website

Universal AI Platform for Industrial Applications

Universal, scalable AI platform for industrial applications. Developing machine learning methods for engineering applications and industrial data analysis in SMEs.

Industrial AI Machine Learning Platform Development
Project Website

Fast Neural Network Research

Research project at The Alan Turing Institute developing fast and efficient neural network approaches for large-scale data analysis and scientific computing applications.

Neural Networks Scientific Computing Efficiency
Project Website

Portfolio

Neural Operator Visualization

Physics-Informed Neural Networks

Implementation of neural operators for solving PDEs in climate modeling with improved computational efficiency.

PyTorch Physics Deep Learning
GNN Visualization

Weather Pattern Recognition

Graph Neural Network approach to identifying and predicting weather patterns from satellite imagery and sensor networks.

GNN Remote Sensing Weather
Climate Data Analysis

Climate Data Analysis Pipeline

End-to-end machine learning pipeline for processing and analyzing large-scale climate datasets.

Data Science Climate Python

Experience & Education

PhD in Machine Learning

UiT The Arctic University of Norway, Tromsø | 2022 - Present

PhD student focusing on AI applications in Earth system science, particularly sea ice forecasting and climate modeling. Contributing to international research projects including IceNet and Visual Intelligence.

Research Assistant

[Research Group/Institution] | 2021 - 2022

Developed machine learning models for environmental data analysis and contributed to publications in climate science journals.

Master's in Physics

University of Duisburg-Essen | 2019 - 2021

Focused on computational physics and data analysis methods. Thesis on applying machine learning to physical systems.

Bachelor's in Physics

University of Duisburg-Essen | 2016 - 2019

Strong foundation in theoretical and experimental physics with emphasis on mathematical modeling and computational methods.

Get In Touch

Let's Collaborate

Interested in discussing research opportunities, collaborations, or potential projects? I'd love to hear from you.

📧 lars.uebbing@uit.no
🏛️ UiT The Arctic University of Norway
📍 Tromsø, Norway