Thomas Doensig Jorgensen

Papers

Total Citations

H-Index

About

Thomas Dønsig Jørgensen is a researcher whose work sits at the intersection of artificial intelligence and information theory, with a particular focus on optimizing neural network architectures. His most cited contribution, "Pruning Artificial Neural Networks Using Neural Complexity Measures" (2008, 26 citations), introduces an innovative method for network compression that leverages information-theoretic complexity to identify and remove redundant connections. This approach offers a principled alternative to heuristic pruning techniques, demonstrating how measures of neural complexity can guide the efficient simplification of networks without sacrificing performance. While his citation count reflects a specialized but impactful niche, Jørgensen’s work is notable for its conceptual depth, bridging theoretical insights from complexity science with practical machine learning challenges. His research provides a foundation for understanding how information-theoretic principles can enhance model efficiency, making it relevant to ongoing efforts in sustainable AI and model optimization. For students and researchers exploring network pruning or complexity-driven learning, Jørgensen’s contributions offer a thoughtful and mathematically grounded perspective.