Data compression system
AssigneeMicrosoft Technology Licensing, LLC
InventorsSebastian NOWOZIN, Ryota TOMIOKA, Diane BOUCHACOURT
Life of this patent
Estimating the term…
Abstract
A data compression apparatus is described which has an encoder configured to receive an input data item and to compress the data item into an encoding comprising a plurality of numerical values. The numerical values are grouped at least according to whether they relate to content of the input data item or style of the input data item. The encoder has been trained using a plurality of groups of training data items grouped according to the content and where training data items within individual ones of the groups vary with respect to the style. The encoder has been trained using a training objective which takes into account the groups.
Claims20
A computer-implemented data compression apparatus comprising:
- an encoder configured to receive an input data item and to compress the data item into an encoding comprising a plurality of numerical values, where the numerical values are grouped at least according to whether they relate to content of the input data item or style of the input data item;
- a memory configured to store the encoding;
- and
- wherein the encoder has been trained using:
- a plurality of groups of training data items grouped according to the content and where training data items within individual ones of the groups vary with respect to the style, and
- a training objective which takes into account the groups.
A computer-implemented data decompression apparatus comprising:
- a decoder configured to receive an encoding of a data item, the encoding comprising a plurality of numerical values, where the numerical values are separated at least according to whether they relate to content of the data item or style of the data item, and to decode the encoding into an output data item which is substantially the same as the data item;
- and
- wherein the decoder has been trained using:
- a plurality of groups of training data items grouped according to the content and where training data items within individual ones of the groups vary with respect to the style, and
- a training objective which takes into account the groups.
A computer-implemented method comprising:
- training, using a processor, an encoder to receive an input data item and to compress the data item into an encoding comprising a plurality of numerical values, where the numerical values are grouped at least according to whether they relate to content of the input data item or style of the input data item;
- storing the encoding;
- and
- wherein the training comprises using:
- a plurality of groups of training data items grouped according to the content and where training data items within individual ones of the groups vary with respect to the style, and
- a training objective which takes into account the groups.
Dependencies are inferred from the claim text.