US 11,385,901 B2Granted

Systems and methods of parallel and distributed processing of datasets for model approximation

AssigneeCapital One Services, LLC

InventorsMatthew van Adelsberg, Rohit Joshi, Siqi Wang

Life of this patent

Estimating the term…

Loading patent…

Abstract

A system including: at least one processor; and at least one memory having stored thereon computer program code that, when executed by the at least one processor, controls the system to: receive a data model identification and a dataset; in response to determining that the data model does not contain a hierarchical structure, perform expectation propagation on the dataset to approximate the data model with a hierarchical structure; divide the dataset into a plurality of channels; for each of the plurality of channels: divide the data into a plurality of microbatches; process each microbatch of the plurality of microbatches through parallel iterators; and process the output of the parallel iterators through single-instruction multiple-data (SIMD) layers; and asynchronously merge results of the SIMD layers.

Claims20

Dependent claims

A system comprising:

at least one hardware processor;
and
at least one hardware memory having stored thereon computer program code that, when executed by the at least one hardware processor, controls the system to:
receive a data model identification and a dataset of a data model;
in response to determining that the data model does not contain a hierarchical structure, perform expectation propagation on the dataset for the data model to approximate the data model for the dataset with an approximate data model having a hierarchical structure;
divide the dataset into a plurality of channels of data, each of the plurality of channels of data comprising a separate logical division of the data;
for each of the plurality of channels of data:
divide the data into a plurality of microbatches, a number of the plurality of microbatches being based on model parameters and sufficiency statistics;
process each microbatch of the plurality of microbatches through parallel iterators;
and
process outputs of the parallel iterators through single-instruction multiple-data (SIMD) layers;
and
asynchronously merge results of the SIMD layers to create merged results.

10 more elements

Dependencies are inferred from the claim text.

Facts

Application: 15/930,728Filed 13 May 2020 · continuation
Priority: 2 May 2019
Published: US 2020/0348941 A15 Nov 2020
Granted: 12 Jul 2022
Term ends: Not estimatedThe source records don’t support an estimate.

Classification8 codes

G06F 9/3887
Using a plurality of independent parallel functional units: controlled by a single instruction for multiple data lanes [SIMD]
G06N 20/00
Machine learning
G06F 9/268
Arrangements for next microinstruction selection: microinstruction selection not based on processing results, e.g. interrupt, patch, first cycle…
G06F 9/4806
Program initiating; Program switching, e.g. by interrupt: task transfer initiation or dispatching