Abstract: The present disclosure is directed to methods of machine-learning for on-the-fly adaptation of a predictive model configured for image processing. The method includes providing a first predictive model to provide first prediction task, providing a second predictive model to provide a second prediction task, the second predictive model including one or more parameters, the second prediction task being derivable from the first prediction task, providing context-based images from a first stream of images obtained from an optical sensor to both the first and the second predictive models, each image provided on-the-fly, and performing an on-the-fly adaptation for the second predictive model, the on-the-fly adaptation comprising, for each provided image: performing a respective first prediction by the first predictive model and a respective second prediction by the second predictive model, computing a cost function, and updating the one or more parameters of the second predictive model based on the cost function.
Type:
Grant
Filed:
March 11, 2022
Date of Patent:
October 27, 2026
Assignee:
Datakalab
Inventors:
Arnaud Dapogny, Kevin Bailly, Lucas Fischer, Lucas Lugao Guimaraes
Abstract: The disclosure notably relates to a computer-implemented method for neural network quantization. The method comprises obtaining a neural network. The method further comprises obtaining a quantization operator. The quantization operator is based on a power function. The power function has a power exponent. The method further comprises quantizing the neural network based on the quantization operator. The quantization includes searching for an optimal value of the power exponent based on a quantization error associated with the quantization operator.
Type:
Application
Filed:
May 3, 2023
Publication date:
November 14, 2024
Applicant:
Datakalab
Inventors:
Edouard Yvinec, Arnaud Dapogny, Kevin Bailly, Lucas Fischer
Abstract: The present invention is notably directed to methods of machine-learning for an on-the-fly adaptation of a predictive model configured for image processing.
Type:
Application
Filed:
March 11, 2022
Publication date:
May 16, 2024
Applicant:
Datakalab
Inventors:
Arnaud Dapogny, Kevin Bailly, Lucas Fischer, Lucas Lugao Guimaraes