Patents by Inventor Po-Yen Hsieh

Po-Yen Hsieh has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20240113071
    Abstract: An integrated circuit package including electrically floating metal lines and a method of forming are provided. The integrated circuit package may include integrated circuit dies, an encapsulant around the integrated circuit dies, a redistribution structure on the encapsulant, a first electrically floating metal line disposed on the redistribution structure, a first electrical component connected to the redistribution structure, and an underfill between the first electrical component and the redistribution structure. A first opening in the underfill may expose a top surface of the first electrically floating metal line.
    Type: Application
    Filed: January 5, 2023
    Publication date: April 4, 2024
    Inventors: Chung-Shi Liu, Mao-Yen Chang, Yu-Chia Lai, Kuo-Lung Pan, Hao-Yi Tsai, Ching-Hua Hsieh, Hsiu-Jen Lin, Po-Yuan Teng, Cheng-Chieh Wu, Jen-Chun Liao
  • Patent number: 11636336
    Abstract: A training device and a training method for a neural network model. The training method includes: obtaining a data set; completing, according to the data set, a plurality of artificial intelligence (AI) model trainings to generate a plurality of models corresponding to the plurality of AI model trainings respectively; selecting, according to a first constraint, a first model set from the plurality of models; and selecting, according to a second constraint, the neural network model from the first model set.
    Type: Grant
    Filed: December 29, 2019
    Date of Patent: April 25, 2023
    Assignee: Industrial Technology Research Institute
    Inventors: Mao-Yu Huang, Po-Yen Hsieh, Chih-Neng Liu, Tsann-Tay Tang
  • Publication number: 20210174200
    Abstract: A training device and a training method for a neural network model are provided. The training method includes: obtaining a data set; completing, according to the data set, a plurality of artificial intelligence (AI) model trainings to generate a plurality of models corresponding to the plurality of AI model trainings respectively; selecting, according to a first constraint, a first model set from the plurality of models; and selecting, according to a second constraint, the neural network model from the first model set.
    Type: Application
    Filed: December 29, 2019
    Publication date: June 10, 2021
    Applicant: Industrial Technology Research Institute
    Inventors: Mao-Yu Huang, Po-Yen Hsieh, Chih-Neng Liu, Tsann-Tay Tang