the Gram-Schmidt process does: orthogonalizes a set of vectors

batch size affects generalization: larger batches find sharper minima

Larger batch sizes lead to sharper minima, enhancing generalization by providing more accurate gradient estimates

Vector quantization

Product quantization compresses vectors by splitting them into subvectors and quantizing each subvector independently

data augmentation does for generalization: artificially expands training set

Data augmentation artificially expands the training set, enhancing model generalization

orthogonal matrices preserve distances: O^T O = I means no stretching or squashing

Orthogonal matrices preserve distances because O^T O = I ensures no stretching or squashing occurs

384-dim all-MiniLM-L6-v2 optimizes: fast sentence similarity with 6 layers

All-MiniLM-L6-v2 optimizes fast sentence similarity with 6 layers

quantization to INT8 doubles throughput

Quantization to INT8 doubles throughput because tensor cores process INT8 2x faster