FAISS Index Types and HNSW Algorithm Explained
FlatL2 - Euclidean Distance For vectors: $$\mathbf{x} = (x_1, x_2, ..., x_d)$$ $$\mathbf{y} = (y_1, y_2, ..., y_d)$$ The squared Euclidean distance (L2 distance squared) is defined as: $$D(\mathbf{x}, \mathbf{y}) = |\mathbf{x} - \mathbf{y}|^2 = \sum_{i=1}^{d} (x_i - y_i)^2$$ The Euclidean distance i...