Scalar Quantization
A quantization method that rounds each numerical value independently, as opposed to vector quantization.
Timeline (last 90 days)
Vector quantization groups multiple values into a vector and maps it as a whole to the nearest codeword in a pre-learned codebook, typically retaining more information than scalar quantization at the same bit rate.
Common approaches to vector storage compression include dimensionality reduction (such as Matryoshka representation learning), scalar quantization (compressing float32 to int8), and product quantization (PQ).
All Facts (2)
Vector quantization groups multiple values into a vector and maps it as a whole to the nearest codeword in a pre-learned codebook, typically retaining more information than scalar quantization at the same bit rate.
50%UnverifiedCommon approaches to vector storage compression include dimensionality reduction (such as Matryoshka representation learning), scalar quantization (compressing float32 to int8), and product quantization (PQ).
50%