Systems and methods for bayesian optimization using integrated acquisition functions

Patent metadata

Field	Value
Publication number	US-2016292129-A1
Application number	US-201414291212-A
Country	US
Kind code	A1
Filing date	May 30, 2014
Priority date	May 30, 2013
Publication date	Oct 6, 2016
Grant date	—

Abstract

Techniques for use in connection with performing optimization using an objective function. The techniques include using at least one computer hardware processor to perform: identifying, using an integrated acquisition utility function and a probabilistic model of the objective function, at least a first point at which to evaluate the objective function; evaluating the objective function at least at the identified first point; and updating the probabilistic model of the objective function using results of the evaluating to obtain an updated probabilistic model of the objective function.

First claim

What is claimed is:

1 . A system for use in connection with performing optimization using an objective function, the system comprising:
at least one computer hardware processor; and at least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by the at least one computer hardware processor, cause the at least one computer hardware processor to perform:
identifying, using an integrated acquisition utility function and a probabilistic model of the objective function, at least a first point at which to evaluate the objective function;
evaluating the objective function at least at the identified first point; and
updating the probabilistic model of the objective function using results of the evaluating to obtain an updated probabilistic model of the objective function.

2 . The system of claim 1 , wherein the objective function relates values of hyper-parameters of a machine learning system to values providing a measure of performance of the machine learning system.

3 . The system of claim 2 , wherein the objective function relates values of a plurality of hyper-parameters of a neural network for identifying objects in images to respective values providing a measure of performance of the neural network in identifying the objects in the images.

4 . The system of claim 1 , wherein the processor-executable instructions further cause the at least one computer hardware processor to perform:
identifying, using the integrated acquisition utility function and the updated probabilistic model of the objective function, at least a second point at which to evaluate the objective function; and evaluating the objective function at least at the identified second point.

5 . The system of claim 1 , wherein the probabilistic model has at least one parameter, and wherein the integrated acquisition utility function is obtained at least in part by integrating an initial acquisition utility function with respect to the at least one parameter of the probabilistic model.

6 . The system of claim 5 , wherein the initial acquisition utility function is an acquisition utility function selected from the group consisting of: a probability of improvement utility function, an expected improvement utility function, a regret minimization utility function, and an entropy-based utility function.

7 . The system of claim 1 , wherein the probabilistic model of the objective function comprises a Gaussian process or a neural network.

8 . The system of claim 1 , wherein the identifying is performed at least in part by using a Markov chain Monte Carlo technique.

9 . The system of claim 1 , wherein the processor-executable instructions further cause the at least one computer hardware processor to perform:
identifying a plurality of points at which to evaluate the objective function; evaluating the objective function at each of the plurality of points; and identifying or approximating, based on results of the evaluating, a point at which the objective function attains a maximum value.

10 . A method for use in connection with performing optimization using an objective function, the method comprising:
using at least one computer hardware processor to perform:
identifying, using an integrated acquisition utility function and a probabilistic model of the objective function, at least a first point at which to evaluate the objective function;
evaluating the objective function at least at the identified first point; and
updating the probabilistic model of the objective function using results of the evaluating to obtain an updated probabilistic model of the objective function.

11 . The method of claim 10 , wherein the objective function relates values of hyper-parameters of a machine learning system to values providing a measure of performance of the machine learning system.

12 . The method of claim 10 , wherein the processor-executable instructions further cause the at least one computer hardware processor to perform:
identifying, using the integrated acquisition utility function and the updated probabilistic model of the objective function, at least a second point at which to evaluate the objective function; and evaluating the objective function at least at the identified second point.

13 . The method of claim 10 , wherein the probabilistic model of the objective function comprises a Gaussian process or a neural network.

14 . The method of claim 10 , wherein the identifying is performed at least in part by using a Markov chain Monte Carlo technique.

15 . The method of claim 10 , wherein the processor-executable instructions further cause the at least one computer hardware processor to perform:
identifying a plurality of points at which to evaluate the objective function; evaluating the objective function at each of the plurality of points; and identifying or approximating, based on results of the evaluating, a point at which the objective function attains a maximum value.

16 . At least one non-transitory computer-readable storage medium storing processor executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a method for use in connection with performing optimization using an objective function, the method comprising:
identifying, using an integrated acquisition utility function and a probabilistic model of the objective function, at least a first point at which to evaluate the objective function; evaluating the objective function at least at the identified first point; and updating the probabilistic model of the objective function using results of the evaluating to obtain an updated probabilistic model of the objective function.

17 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the objective function relates values of hyper-parameters of a machine learning system to values providing a measure of performance of the machine learning system.

18 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the processor-executable instructions further cause the at least one computer hardware processor to perform:
identifying, using the integrated acquisition utility function and the updated probabilistic model of the objective function, at least a second point at which to evaluate the objective function; and evaluating the objective function at least at the identified second point.

19 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the probabilistic model of the objective function comprises a Gaussian process or a neural network.

20 . The at least one non-transitory computer-readable storage medium of claim 16 , wherein the identifying is performed at least in part by using a Markov chain Monte Carlo technique.

Assignees

• Univ Sherbrooke
• Harvard College
• Governing Council Of The Univ Of Toronto The

Inventors

• Adams Ryan P
• Snoek Roland Jasper
• Larochelle Hugo

Classifications

• G06N7/01 Primary

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• G06N20/10 Primary

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06F17/11 Primary

  for solving equations {, e.g. nonlinear equations, general mathematical optimization problems (optimization specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G06N20/00

  Machine learning · CPC title

• G06N5/048 Primary

  Fuzzy inferencing · CPC title