# US 2023/0409981 A1 · Kernel learning apparatus using transformed convex optimization problem

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0409981 A1
- Document ID: US20230409981A1
- Status: Published application
- Kind: A1
- Application number: 18/240,221
- Filed: 2023-08-30
- Published: 2023-12-21
- Earliest claimed: 2020-09-25
- Assignees: NEC Corporation
- Applicants: NEC Corporation
- Inventors: Hao ZHANG; Shinji NAKADAI; Kenji FUKUMIZU
- Web page: https://patentgrep.com/patent/US20230409981A1

## CPC

- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06F 17/14: Fourier, Walsh or analogous domain transformations {, e.g. Laplace, Hilbert, Karhunen-Loeve, transforms (for correlation function computation G06F17/156; spectrum analysers G01R23/16)}
- G06F 18/213: Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F 18/21355: {nonlinear criteria, e.g. embedding a manifold in a Euclidean space}
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

In a kernel learning apparatus, a data preprocessing circuitry preprocesses and represents each data example as a collection of feature representations that need to be interpreted. An explicit feature mapping circuit designs a kernel function with an explicit feature map to embed the feature representations of data into a nonlinear feature space and to produce the explicit feature map for the designed kernel function to train a predictive model. A convex problem formulating circuitry formulates a non-convex problem for training the predictive model into a convex optimization problem based on the explicit feature map. An optimal solution solving circuitry solves the convex optimization problem to obtain a globally optimal solution for training an interpretable predictive model.

## Claims

### Claim 1

A kernel learning device comprising:
an explicit feature mapping circuit configured to design a kernel function with an explicit feature map to embed feature representations of data into a nonlinear feature space and to produce the explicit feature map of the designed kernel function to train a machine learning model;
a convex problem formulating circuit configured to formulate a non-convex problem for training the machine learning model into a convex optimization problem based on the explicit feature map;
an optimal solution solving circuit configured to solve the convex optimization problem to obtain an optimal solution for training an interpretable machine learning model; and
an output part configured to output a graph obtained by supplying dataset to the trained interpretable machine learning model.

### Claim 2 (depends on 1)

The kernel learning device according to claim 1, wherein the output part produces, as the graph, degree of importance for features in a machine-learning task.

### Claim 3 (depends on 1)

The kernel leaning device according to claim 1, wherein the graph illustrates a relationship between values of each feature and contribution for a machine-learned value of the feature.

### Claim 4 (depends on 3)

The kernel leaning device according to claim 3, wherein the graph represents partial dependence of the contribution.

### Claim 5 (depends on 3)

The kernel leaning device according to claim 3, wherein the graph denotes the partial dependence at a change of shading in a color.
