公开(公告)号：US11605020B2

公开(公告)日：2023-03-14

申请号：US16519914

申请日：2019-07-23

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Changchuan Yin

IPC: G06F40/103 ,  G06N20/00

Abstract: A processing system including at least one processor may obtain a machine learning model, serialize the machine learning model into a serialized format, and embed a delimiter indicator into a documentation file comprising information regarding the use of the machine learning model, where the delimiter indicator is in a file position that is after an end-of-file indicator of the documentation file. The processing system may further embed the machine learning model in the serialized format into the documentation file in a file position that is after the delimiter indicator. The processing system may then store the documentation file with the delimiter indicator and the machine learning model in the serialized format that are embedded.

公开(公告)号：US11361159B2

公开(公告)日：2022-06-14

申请号：US17241361

申请日：2021-04-27

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Changchuan Yin

IPC: G06F40/284 ,  G06F40/242

Abstract: Text can be encoded into DNA sequences. Each word from a document or other text sample can be encoded in a DNA sequence or DNA sequences and the DNA sequences can be stored for later retrieval. The DNA sequences can be stored digitally, or actual DNA molecules containing the sequences can be synthesized and stored. In one example, the encoding technique makes use of a polynomial function to transform words based on the Latin alphabet into k-mer DNA sequences of length k. Because the whole bits required for the DNA sequences are smaller than the actual strings of words, storing documents using DNA sequences may compress the documents relative to storing the same documents using other techniques. In at least one example, the mapping between words and DNA sequences is one-to-one and the collision ratio for the encoding is low.

公开(公告)号：US20210366170A1

公开(公告)日：2021-11-25

申请号：US16880503

申请日：2020-05-21

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Changchuan Yin

IPC: G06T11/20 ,  G06F40/126 ,  G06T11/00

Abstract: Aspects of the subject disclosure may include, for example, a method of transforming, by a processing system comprising a processor, text comprising a series of characters into a graphic representation, wherein the graphic representation comprises a series of dots arranged in a two-dimensional pattern, wherein the two-dimensional pattern comprises two dots per character, and wherein each dot in the series of dots is one unit away from a preceding dot; and plotting, by the processing system, the series of dots on a two-dimensional graph, thereby creating a unique encoded image of the text. Other embodiments are disclosed.

公开(公告)号：US11017170B2

公开(公告)日：2021-05-25

申请号：US16143671

申请日：2018-09-27

Applicant: AT&T Intellectual Property I, L.P.

Inventor： Changchuan Yin

IPC: G06F40/284 ,  G06F40/242

Abstract: Text can be encoded into DNA sequences. Each word from a document or other text sample can be encoded in a DNA sequence or DNA sequences and the DNA sequences can be stored for later retrieval. The DNA sequences can be stored digitally, or actual DNA molecules containing the sequences can be synthesized and stored. In one example, the encoding technique makes use of a polynomial function to transform words based on the Latin alphabet into k-mer DNA sequences of length k. Because the whole bits required for the DNA sequences are smaller than the actual strings of words, storing documents using DNA sequences may compress the documents relative to storing the same documents using other techniques. In at least one example, the mapping between words and DNA sequences is one-to-one and the collision ratio for the encoding is low.