intrinsic geometry
In AI, intrinsic geometry refers to the geometric properties of a data space that are inherent to the distribution of the data points, rather than defined by external metrics. Understanding intrinsic geometry can be crucial in tasks like manifold learning where the goal is to discover low-dimensional structures within high-dimensional data.
- A geometric framework for momentum-based optimizers for low-rank training
- Follow the Energy, Find the Path: Riemannian Metrics from Energy-Based Models
- Fréchet Geodesic Boosting
- Learning Generalizable Shape Completion with SIM(3) Equivariance
- ShapeEmbed: a self-supervised learning framework for 2D contour quantification
- ViSPLA: Visual Iterative Self-Prompting for Language-Guided 3D Affordance Learning