VC86/QRE-QEP

Implements quantized random embeddings for dimensionality reduction of disjoint convex sets, offering an alternative to PCA with theoretical guarantees via the restricted isometry property. The codebase provides numerical experiments comparing quantization-based approaches against traditional adaptive methods on supervised learning tasks. Focuses on preserving geometric structure of low-complexity sets while operating in quantized (discrete) embedding spaces.

No commits in the last 6 months.