If in Singapore next week, come by our #ICLR2025 Spotlight poster for our recent study at @KempeLab unveiling how data pruning promotes implicit bias in datasets and proposing a method (DRoP) that does exactly the opposite:

Check out our recent work with Julia Kempe (@KempeLab) and Anna Dawid (@MolecularRobot) on the properties of the loss Hessian eigenspectrum across parameter scales, new type of vanishing gradient, and mysterious GD dynamics - to be presented at #ICML2024 -