Neural network validation system

Patent metadata

Field	Value
Publication number	US-2023139521-A1
Application number	US-202117517260-A
Country	US
Kind code	A1
Filing date	Nov 2, 2021
Priority date	Nov 2, 2021
Publication date	May 4, 2023
Grant date	—

Abstract

A system comprises a computer including a processor and a memory. The memory includes instructions such that the processor is programmed to: receive, at a first neural network, unlabeled sensor data, wherein the first neural network generates output based on the unlabeled sensor data, receive, at a second neural network, the unlabeled sensor data, wherein the second neural network generates output based on the unlabeled sensor data during a validation mode, the second neural network different from the first neural network, compare the output generated by the first neural network with the output generated by the second neural network, and generate an alert when a difference between the output generated by the first neural network and the output generated by the second neural network is greater than a predetermined comparison threshold.

First claim

What is claimed is:

1 . A system comprising a computer including a processor and a memory, the memory including instructions such that the processor is programmed to:
receive, at a first neural network, unlabeled sensor data, wherein the first neural network generates output based on the unlabeled sensor data; receive, at a second neural network, the unlabeled sensor data, wherein the second neural network generates output based on the unlabeled sensor data during a validation mode, the second neural network different from the first neural network; compare the output generated by the first neural network with the output generated by the second neural network; and generate an alert when a difference between the output generated by the first neural network and the output generated by the second neural network is greater than a predetermined comparison threshold.

2 . The system of claim 1 , wherein the processor is further programmed to receive a selection to transition between the validation mode and a feature mode.

3 . The system of claim 2 , wherein the processor is further programmed to operate at least one vehicle actuator based on the output generated by the first neural network during the feature mode.

4 . The system of claim 2 , wherein the selection is transmitted from a server.

5 . The system of claim 2 , wherein the selection is transmitted from an electronic controller unit of a vehicle.

6 . The system of claim 1 , wherein the first neural network is trained using a first dataset and the second neural network is trained using a second dataset, wherein the second dataset is different from the first dataset.

7 . The system of claim 1 , wherein the processor is further programmed to prevent the output generated by the first neural network from being used to operate a vehicle during the validation mode.

8 . The system of claim 1 , wherein the unlabeled sensor data comprises sensor data collected by a fleet of vehicles.

9 . A vehicle including a system, the system comprising a computer including a processor and a memory, the memory including instructions such that the processor is programmed to:
receive, at a first neural network, unlabeled sensor data, wherein the first neural network generates output based on the unlabeled sensor data; receive, at a second neural network, the unlabeled sensor data, wherein the second neural network generates output based on the unlabeled sensor data during a validation mode, the second neural network different from the first neural network; compare the output generated by the first neural network with the output generated by the second neural network; and generate an alert when a difference between the output generated by the first neural network and the output generated by the second neural network is greater than a predetermined comparison threshold.

10 . The vehicle of claim 9 , wherein the processor is further programmed to receive a selection to transition between the validation mode and a feature mode.

11 . The vehicle of claim 10 , wherein the processor is further programmed to operate at least one vehicle actuator of the vehicle based on the output generated by the first neural network during the feature mode.

12 . The vehicle of claim 10 , wherein the selection is transmitted from a server.

13 . The system of claim 10 , wherein the selection is transmitted from an electronic controller unit of the vehicle.

14 . The system of claim 9 , wherein the first neural network is trained using a first dataset and the second neural network is trained using a second dataset, wherein the second dataset is different from the first dataset.

15 . The system of claim 9 , wherein the processor is further programmed to prevent the output generated by the first neural network from being used to operate a vehicle during the validation mode.

16 . The system of claim 9 , wherein the unlabeled sensor data comprises sensor data collected by a fleet of vehicles.

17 . A method comprising:
receiving, at a first neural network, unlabeled sensor data, wherein the first neural network generates output based on the unlabeled sensor data; receiving, at a second neural network, the unlabeled sensor data, wherein the second neural network generates output based on the unlabeled sensor data during a validation mode, the second neural network different from the first neural network; comparing the output generated by the first neural network with the output generated by the second neural network; and generating an alert when a difference between the output generated by the first neural network and the output generated by the second neural network is greater than a predetermined comparison threshold.

18 . The method of claim 17 , further comprising receiving a selection to transition between the validation mode and a feature mode.

19 . The method of claim 18 , further comprising operating at least one vehicle actuator based on the output generated by the first neural network during the feature mode.

20 . The system of claim 18 , wherein the selection is transmitted from a server.

Assignees

• Gm Global Tech Operations Llc

Inventors

• Tong Wei
• Wang Shige
• Sethu Ramesh
• Scheu Jeffrey D
• Radhakrishan Prashanth
• Mudalige Upali P
• Ahmed Ryan

Classifications

• G05D1/0291 Primary

  Fleet control (monitoring fleets in traffic control systems for road vehicles G08G1/127, G08G1/127) · CPC title

• G06N3/044

  Recurrent networks, e.g. Hopfield networks · CPC title

• G06F18/2193

  based on specific statistical tests · CPC title

• G06F18/2155 Primary

  characterised by the incorporation of unlabelled data, e.g. multiple instance learning [MIL], semi-supervised techniques using expectation-maximisation [EM] or naïve labelling · CPC title

• G06N3/045

  Combinations of networks · CPC title