I'm a computer science student at TU Darmstadt working on machine learning research. My work focuses on improving model training across various settings.
Selected work:
- Improved Elastic Weight Consolidation — an optimization-theoretic derivation of continual-learning regularization [2026] Code Paper
- Temporal Regularized Learning — self-supervised learning local in space and time [2026] Code Paper
- pptrain — task-agnostic training on synthetic and procedural tasks [2026] Code
- Closed-Form Initialization — exploring whether neural networks, including transformers, can be initialized analytically [2026] Code
- Gradient Routing for Continual Learning — testing whether sparsity can reduce catastrophic forgetting [2026] Code
Reach me at: davide.wiest2 [at] gmail [dot] com