Canon kabushiki kaisha (20240281678). INFERENCE APPARATUS, IMAGING APPARATUS, METHOD OF CONTROLLING INFERENCE APPARATUS, AND STORAGE MEDIUM simplified abstract
INFERENCE APPARATUS, IMAGING APPARATUS, METHOD OF CONTROLLING INFERENCE APPARATUS, AND STORAGE MEDIUM
Organization Name
Inventor(s)
REIJI Hasegawa of Kanagawa (JP)
INFERENCE APPARATUS, IMAGING APPARATUS, METHOD OF CONTROLLING INFERENCE APPARATUS, AND STORAGE MEDIUM - A simplified explanation of the abstract
This abstract first appeared for US patent application 20240281678 titled 'INFERENCE APPARATUS, IMAGING APPARATUS, METHOD OF CONTROLLING INFERENCE APPARATUS, AND STORAGE MEDIUM
The abstract describes an inference apparatus that utilizes machine learned models to output likelihoods of different classes and perform computations in parallel to improve efficiency.
- The apparatus includes a first machine learned model with multiple learners and a plurality of second machine learned models corresponding to different classes.
- Computation of the second machine learned model corresponding to a selected class is done simultaneously with the remaining computation of the first machine learned model.
- If the selected class coincides with the first class after the first model's computation is completed, the computation of the second model continues for that class.
- This process results in an inference result based on the likelihoods calculated by the machine learned models.
Potential Applications: - This technology can be applied in various fields such as image recognition, natural language processing, and predictive analytics. - It can enhance the accuracy and efficiency of classification tasks in machine learning systems.
Problems Solved: - Improves the speed and accuracy of inference tasks by utilizing parallel computation. - Enhances the overall performance of machine learning models by optimizing the computation process.
Benefits: - Faster and more accurate inference results. - Improved efficiency in processing large datasets. - Enhanced performance of machine learning models.
Commercial Applications: Title: Enhanced Machine Learning Inference Apparatus for Efficient Classification This technology can be utilized in industries such as healthcare, finance, and e-commerce for tasks like medical diagnosis, fraud detection, and personalized recommendations.
Questions about the technology: 1. How does this technology improve the efficiency of machine learning inference tasks? - The technology utilizes parallel computation to speed up the process and enhance accuracy. 2. What are the potential real-world applications of this enhanced inference apparatus? - The technology can be applied in various industries for tasks requiring classification and prediction.
Original Abstract Submitted
an inference apparatus includes a first machine learned model including a plurality of learners and outputting likelihoods of a plurality of classes, and a plurality of second machine learned models corresponding to the plurality of classes, performs computation of a second machine learned model corresponding to a first class selected from the plurality of classes based on likelihoods calculated in the middle of computation of the plurality of learners in the first machine learned model, in parallel with remaining computation of the plurality of learners in the first machine learned model, and in a case where a second class selected based on likelihoods when computation of the plurality of learners in the first machine learned model is fully completed is coincident with the first class, continues the computation of the second machine learned model corresponding to the first class, thereby outputting an inference result.
