IP Library › Granted Patent US 12,633,105
Granted Patent B2
US 12,633,105 · App. 18/473,136 · Granted May 19, 2026

Automatically ascertaining an optimal architecture for a neural network

Inventors: Benedikt Sebastian Staffler (Tuebingen, DE); David Stoeckel (Rutesheim, DE); Thomas Elsken (Vrees, DE)
Assignee: ROBERT BOSCH GMBH
G06V10/82G06V10/776G06V10/84G06V10/87
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,633,105
App. No.
18/473,136
Granted
May 19, 2026
Kind
B2
Abstract

A computer-implemented method for ascertaining an optimal architecture for a neural network that solves a given task in accordance with given boundary conditions and/or optimization goals. The method includes: providing a graph of the possible architectures of nodes and edges, wherein nodes correspond to data, edges correspond to parameterized operations to be carried out on the data, and a path which traverses the entire graph corresponds to an architecture; in a search phase, generating candidate architectures based on already known architectures, wherein the candidate architectures are similar but not identical to the known architectures in accordance with a predetermined criterion; evaluating the candidate architectures using the given boundary conditions and/or optimization goals; ascertaining a candidate architecture having the best rating as the sought optimal architecture.

Claims (49)

1 . A computer-implemented method for ascertaining an optimal architecture for a neural network that solves a given task in accordance with given boundary conditions and/or optimization goals, the method comprising the following steps:

providing a graph of possible architectures of nodes and edges, wherein;

each of the nodes corresponds to data;

each of the edges corresponds to one or more parameterized operations to be carried out on the data; and

a path which traverses the entire graph corresponds to an architecture of the possible architectures;

generating, in a search phase, candidate architectures based on already known architectures, wherein the candidate architectures are similar but not identical to the already known architectures in accordance with a predetermined criterion;

evaluating the candidate architectures using the given boundary conditions and/or optimization goals;

ascertaining a candidate architecture of the candidate architectures having a best rating as the ascertained optimal architecture;

providing the ascertained optimal architecture with measurement data recorded with at least one sensor;

ascertaining a control signal from output obtained from the ascertained optimal architecture; and

controlling, using the control signal, a vehicle, and/or a robot, and/or a driving assistance system, and/or a system for quality control, and/or a system for monitoring areas and/or a system for medical imaging.

2 . The method according to claim 1 , wherein an already known architecture of the already known architectures is modified to at least one candidate architecture by directly changing at least one edge of the already known architecture.

3 . The method according to claim 1 , wherein the boundary conditions and/or optimization goals include:

how quickly, and/or with what accuracy, the candidate architecture solves the given task; and/or

to what extent predetermined types of errors are avoided when solving the given task; and/or

how much working memory is needed to execute the candidate architecture; and/or

how many processors and/or GPUs are needed to execute the candidate architecture.

4 . The method according to claim 1 , wherein fulfilment of multiple boundary conditions and/or optimization goals is aggregated to a rating number.

5 . The method according to claim 1 , wherein the generating of the candidate architectures includes selecting a subset of nodes of the path that corresponds to an already known architecture of the already known architectures and generating a modification of the path that corresponds to the already known architecture that still passes through all of the selected subset of nodes.

6 . The method according to claim 5 , wherein the generating of the candidate architectures includes generating a modification of the path that corresponds to the already known architecture, which, in accordance with a predetermined metric, extends at no more than a predetermined distance from the path that corresponds to the already known architecture.

7 . The method according to claim 1 , wherein the generating of the candidate architectures includes sampling paths through the graph from a probability distribution that maximizes proximity to and/or intersections with the path that corresponds to an already known architecture of the already known architectures.

8 . The method according to claim 1 , wherein, as part of the evaluation using the boundary conditions and/or optimization goals, a candidate architecture of the candidate architectures is provided with unseen test data and/or validation data in a training phase.

9 . The method according to claim 1 , wherein the given task includes evaluating measurement data in the form of image data and/or point clouds.

10 . The method according to claim 1 , wherein the given task includes mapping input of the neural network to classification scores relating to one or more classes of a predetermined classification and/or to a regression value relating to at least one quantity of interest.

11 . The method according to claim 1 , wherein the candidate architectures are generated as part of an evolutionary algorithm.

12 . The method according to claim 1 , wherein, before the search phase, learning steps regarding the given task are carried out in a training phase for architectures sampled from the graph.

13 . The method according to claim 1 , wherein the ascertained optimal architecture is trained or further trained with training examples respectively labeled with target outputs relating to the given task.

14 . A non-transitory machine-readable storage medium on which is stored one or more computer programs for ascertaining an optimal architecture for a neural network that solves a given task in accordance with given boundary conditions and/or optimization goals, the one or more computer programs, when executed by one or more processors, causing the one or more processors to perform the following steps:

providing a graph of possible architectures of nodes and edges, wherein:

each of the nodes corresponds to data;

each of the edges corresponds to one or more parameterized operations to be carried out on the data; and

a path which traverses the entire graph corresponds to an architecture of the possible architectures;

generating, in a search phase, candidate architectures based on already known architectures, wherein the candidate architectures are similar but not identical to the already known architectures in accordance with a predetermined criterion;

evaluating the candidate architectures using the given boundary conditions and/or optimization goals;

ascertaining a candidate architecture of the candidate architectures having a best rating as the ascertained optimal architecture;

providing the ascertained optimal architecture with measurement data recorded with at least one sensor;

ascertaining a control signal from output obtained from the ascertained optimal architecture; and

controlling, using the control signal, a vehicle, and/or a robot, and/or a driving assistance system, and/or a system for quality control, and/or a system for monitoring areas and/or a system for medical imaging.

15 . One or more computers comprising one or more processors and a non-transitory machine-readable storage medium on which is stored one or more computer programs for ascertaining an optimal architecture for a neural network that solves a given task in accordance with given boundary conditions and/or optimization goals, the one or more computer programs, when executed by the one or more processors, causing the one or more processors to perform the following steps:

providing a graph of possible architectures of nodes and edges, wherein:

each of the nodes corresponds to data;

each of the edges corresponds to one or more parameterized operations to be carried out on the data, and

a path which traverses the entire graph corresponds to an architecture of the possible architectures;

generating, in a search phase, candidate architectures based on already known architectures, wherein the candidate architectures are similar but not identical to the already known architectures in accordance with a predetermined criterion

evaluating the candidate architectures using the given boundary conditions and/or optimization goals;

ascertaining a candidate architecture of the candidate architectures having a best rating as the ascertained optimal architecture;

providing the ascertained optimal architecture with measurement data recorded with at least one sensor;

ascertaining a control signal from output obtained from the ascertained optimal architecture; and

controlling, using the control signal, a vehicle, and/or a robot, and/or a driving assistance system, and/or a system for quality control, and/or a system for monitoring areas and/or a system for medical imaging.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 21, 2023
From: STAFFLER, BENEDIKT SEBASTIAN; STOECKEL, DAVID; ELSKEN, THOMAS
To: ROBERT BOSCH GMBH
Reel/Frame 065933/0258 →
Priority Claims (1)
DE 10 2022 212 901.6 · Nov 30, 2022 · national
Continuity (1)
Related Publication 20240177471A1 · May 30, 2024
References Cited (28)
US 11604992B2 · Fusi · 2023 [cited by examiner]
US 11727277B2 · Hutter · 2023 [cited by examiner]
US 11836595B1 · Yang · 2023 [cited by examiner]
US 12086214B2 · Staffler · 2024 [cited by examiner]
US 12346817B2 · Zoph · 2025 [cited by examiner]
US 12393840B2 · So · 2025 [cited by examiner]
US 20180336453A1 · Merity · 2018 [cited by examiner]
US 20210201526A1 · Moloney et al. · 2021 [cited by applicant]
US 20220019890A1 · Staffler · 2022 [cited by examiner]
US 20220051079A1 · Laszlo · 2022 [cited by examiner]
US 20220114446A1 · Zela · 2022 [cited by examiner]
US 20220121906A1 · Kokiopoulou · 2022 [cited by examiner]
US 20230022777A1 · Staffler · 2023 [cited by examiner]
US 20230306265A1 · Stoll · 2023 [cited by examiner]
US 20240013026A1 · Metzen · 2024 [cited by examiner]
US 20240296357A1 · Staffler · 2024 [cited by examiner]
US 20240378458A1 · Metzen · 2024 [cited by examiner]
CA 3005241A1 · 2018 [cited by examiner]
WO WO2018156942A1 · 2018 [cited by examiner]
WO WO2019081705A1 · 2019 [cited by examiner]
WO WO2019084560A1 · 2019 [cited by examiner]
WO WO2020160252A1 · 2020 [cited by examiner]
WO WO2020259502A1 · 2020 [cited by examiner]
Karagiannakos, “Neural Architecture Search (NAS): basic principles and different approaches” (pp. 1-18) (Year: 2022). [cited by examiner]
Elsken et al., 2019, “Neural Architecture Search: a Survey” (pp. 1-21) (Year: 2019). [cited by examiner]
Cai et al., “Once-For-All: Train One Network and Specialize It for Efficient Deployment,” International Conference on Learning Representations, 2020, pp. 1-15. <https://openreview.net/pdf?id=HylxE1HKwS> Downloaded Sep. … [cited by applicant]
He et al., “AUTOML: a Survey of the State-Of-The-Art,” Knowledge-Based Systems, vol. 212, 2021, pp. 1-37. [cited by applicant]
Stamoulis et al., “Hyperpower: Power- and Memory-Constrained Hyper-Parameter Optimization for Neural Networks,” 2018 Design, Automation & Test in Europe Conference & Exhibition (Date), 2018, pp. 19-24. [cited by applicant]