Patents by Inventor Zichao Wang

Zichao Wang 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).

  • Patent number: 12159694
    Abstract: A machine learning framework is described for performing generation of candidate molecules for, e.g., drug discovery or other applications. The framework utilizes a pre-trained encoder-decoder model to interface between representations of molecules and embeddings for those molecules in a latent space. A fusion module is located between the encoder and decoder and is used to fuse an embedding for an input molecule with embeddings for one or more exemplary molecules selected from a database that is constructed according to a design criteria. The fused embedding is decoded using the decoder to generate a candidate molecule. The fusion module is trained to reconstruct a nearest neighbor to the input molecule from the database based on the sample of exemplary molecules. An iterative approach may be used during inference to dynamically update the database to include newly generated candidate molecules.
    Type: Grant
    Filed: July 17, 2023
    Date of Patent: December 3, 2024
    Assignee: NVIDIA Corporation
    Inventors: Weili Nie, Zichao Wang, Chaowei Xiao, Animashree Anandkumar
  • Patent number: 12147407
    Abstract: A method for processing formulae includes encoding a formula by: training, with a server, a model by using a machine learning algorithm with a data set that includes a plurality of formulae; transforming, with a processor, a first formula into a tree format using the trained model; converting, with the processor, the tree format of the first formula into a plurality of lists; and encoding, with the processor, the plurality of lists into a fixed dimension vector by leveraging a stacked attention module; and generating one or more formula candidates by: obtaining, with the processor, input information; and generating, with the processor, one or more second formula candidates based on input information by using the stacked attention module with a tree beam search algorithm.
    Type: Grant
    Filed: April 21, 2023
    Date of Patent: November 19, 2024
    Assignees: William Marsh Rice University, University of Massachusetts
    Inventors: Zichao Wang, Shiting Lan, Richard G. Baraniuk
  • Patent number: 12087481
    Abstract: Provided are an auxiliary alloy casting piece, a high-remanence and high-coercive force NdFeB permanent magnet, and preparation methods thereof. The method for preparing the auxiliary alloy casting piece includes the following steps: providing an auxiliary alloy material including, by mass percentage, 40% to 45% of Pr, 1% to 2% of Co, 0.5% to 1% of Ga, 0.6% to 0.8% of B, 0.1% to 0.2% of V, 0.3% to 0.7% of Ti, and a balance of Fe; smelting the auxiliary alloy material to obtain a smelted material; and subjecting the smelted material to a quick-setting casting to obtain the auxiliary alloy casting piece; where the quick-setting casting includes a refining and a casting in sequence.
    Type: Grant
    Filed: December 30, 2022
    Date of Patent: September 10, 2024
    Assignee: Baotou Jinshan Magnetic Material Co., Ltd.
    Inventors: Feng Xia, Yulong Fu, Chen Chen, Hailong Zheng, Zichao Wang, Yonghong Liu, Caina Sun, Yu Wang
  • Publication number: 20240127993
    Abstract: Provided are an auxiliary alloy casting piece, a high-remanence and high-coercive force NdFeB permanent magnet, and preparation methods thereof. The method for preparing the auxiliary alloy casting piece includes the following steps: providing an auxiliary alloy material including, by mass percentage, 40% to 45% of Pr, 1% to 2% of Co, 0.5% to 1% of Ga, 0.6% to 0.8% of B, 0.1% to 0.2% of V, 0.3% to 0.7% of Ti, and a balance of Fe; smelting the auxiliary alloy material to obtain a smelted material; and subjecting the smelted material to a quick-setting casting to obtain the auxiliary alloy casting piece; where the quick-setting casting includes a refining and a casting in sequence.
    Type: Application
    Filed: December 30, 2022
    Publication date: April 18, 2024
    Inventors: Feng XIA, Yulong FU, Chen CHEN, Hailong ZHENG, Zichao WANG, Yonghong LIU, Caina SUN, Yu WANG
  • Publication number: 20240029836
    Abstract: A machine learning framework is described for performing generation of candidate molecules for, e.g., drug discovery or other applications. The framework utilizes a pre-trained encoder-decoder model to interface between representations of molecules and embeddings for those molecules in a latent space. A fusion module is located between the encoder and decoder and is used to fuse an embedding for an input molecule with embeddings for one or more exemplary molecules selected from a database that is constructed according to a design criteria. The fused embedding is decoded using the decoder to generate a candidate molecule. The fusion module is trained to reconstruct a nearest neighbor to the input molecule from the database based on the sample of exemplary molecules. An iterative approach may be used during inference to dynamically update the database to include newly generated candidate molecules.
    Type: Application
    Filed: July 17, 2023
    Publication date: January 25, 2024
    Inventors: Weili Nie, Zichao Wang, Chaowei Xiao, Animashree Anandkumar
  • Publication number: 20230342348
    Abstract: A method for processing formulae includes encoding a formula by: training, with a server, a model by using a machine learning algorithm with a data set that includes a plurality of formulae; transforming, with a processor, a first formula into a tree format using the trained model; converting, with the processor, the tree format of the first formula into a plurality of lists; and encoding, with the processor, the plurality of lists into a fixed dimension vector by leveraging a stacked attention module; and generating one or more formula candidates by: obtaining, with the processor, input information; and generating, with the processor, one or more second formula candidates based on input information by using the stacked attention module with a tree beam search algorithm.
    Type: Application
    Filed: April 21, 2023
    Publication date: October 26, 2023
    Applicants: William Marsh Rice University, University of Massachusetts, Amherst
    Inventors: Zichao Wang, Shiting Lan, Richard G. Baraniuk