Fast efficient vocabulary computation with hashed vocabularies
AssigneeGoogle Inc.
InventorJay Yagnik
Life of this patent
Estimating the term…
Abstract
The disclosed embodiments describe a method, an apparatus, an application specific integrated circuit, and a server that provides a fast and efficient look up for data analysis. The apparatus and server may be configured to obtain data segments from a plurality of input devices. The data segments may be individual unique subsets of the entire data set obtained by a plurality input devices. A hash function may be applied to an aggregated set of the data segments. A result of the hash function may be stored in a data structure. A codebook may be generated from the hash function results.
Claims16
A method comprising:
- obtaining, by a processor in a device, a data segment from a data set, wherein the data segment is a subset of the data set;
- applying a hash function to the data segment;
- storing, by the processor, a result of the hash function as a hash index value, wherein the hash index value is used to generate a vocabulary word in a codebook;
- reducing the number of hash index values in the stored codebook, reducing the number of hash index values comprising:
- combining the hash index value and an additional hash index value into a hash conjunction by executing outer product algorithms on the hash index values, the hash conjunction used as the vocabulary word in the codebook.
An apparatus, comprising:
- an image sensor;
- a microphone,
- a display device;
- a memory device;
- and
- a processor, configured to:
- obtain a data segment from a data set, wherein the data segment is a subset of the data set;
- apply a hash function to the data segment;
- store a result of the hash function as a hash index value, wherein the hash index value is used to generate a vocabulary word in a codebook;
- and
- reduce the number of hash index values in the stored codebook, wherein in reducing the number of hash index values, the processor is configured to:
- combine the hash index value and an additional hash index value into a hash conjunction by executing outer product algorithms on the hash index values, the hash conjunction used as the vocabulary word in the codebook.
Dependencies are inferred from the claim text.