Pre-trained embeddings for programming languages using specialized vocabulary

Building pre-trained contextual embeddings for programming languages using specialized vocabulary Download PDF

Info

Publication number US11429352B2 US11429352B2 US16/917,967 US202016917967A US11429352B2 US 11429352 B2 US11429352 B2 US 11429352B2 US 202016917967 A US202016917967 A US 202016917967A US 11429352 B2 US11429352 B2 US 11429352B2 Authority US United States Prior art keywords programming code embeddings contextual programming trained Prior art date 2020-07-01 Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.) Active, expires 2040-10-21 Application number US16/917,967 Other versions US20220004365A1 (en Inventor Saurabh Pujar Luca Buratti Alessandro Morari Jim Alain Laredo Alfio Massimiliano Gliozzo Gaetano Rossiello Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.) International Business Machines Corp Original Assignee International Business Machines Corp Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.) 2020-07-01 Filing date 2020-07-01 Publication date 2022-08-30 2020-07-01 Application filed by International Business Machines Corp filed Critical International Business Machines Corp 2020-07-01 Priority to US16/917,967 priority Critical patent/US11429352B2/en 2020-07-01 Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION reassignment INTERNATIONAL BUSINESS MACHINES CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: GLIOZZO, ALFIO MASSIMILIANO, LAREDO, JIM ALAIN, MORARI, ALESSANDRO, BURATTI, Luca, PUJAR, SAURABH, ROSSIELLO, GAETANO 2022-01-06 Publication of US20220004365A1 publication Critical patent/US20220004365A1/en 2022-08-30 Application granted granted Critical 2022-08-30 Publication of US11429352B2 publication Critical patent/US11429352B2/en Status Active legal-status Critical Current 2040-10-21 Adjusted expiration legal-status Critical

Links

• USPTO
• USPTO PatentCenter
• USPTO Assignment
• Espacenet
• Global Dossier
• Discuss

• 238000000034 method Methods 0.000 claims abstract description 43
• 238000004590 computer program Methods 0.000 claims abstract description 17
• 238000003860 storage Methods 0.000 claims description 43
• 238000010801 machine learning Methods 0.000 claims description 27
• 238000003058 natural language processing Methods 0.000 claims description 24
• 230000015654 memory Effects 0.000 claims description 14
• 239000013598 vector Substances 0.000 claims description 10
• 238000012549 training Methods 0.000 description 44
• 238000012545 processing Methods 0.000 description 17
• 238000010586 diagram Methods 0.000 description 14
• 230000008569 process Effects 0.000 description 12
• 230000006870 function Effects 0.000 description 11
• 239000010410 layer Substances 0.000 description 8
• 238000004891 communication Methods 0.000 description 7
• 230000005540 biological transmission Effects 0.000 description 5
• 238000007726 management method Methods 0.000 description 5
• 238000004458 analytical method Methods 0.000 description 4
• 238000013459 approach Methods 0.000 description 4
• 238000013461 design Methods 0.000 description 4
• 238000012986 modification Methods 0.000 description 4
• 230000004048 modification Effects 0.000 description 4
• 230000008901 benefit Effects 0.000 description 3
• 238000005516 engineering process Methods 0.000 description 3
• 230000003993 interaction Effects 0.000 description 3
• 230000003287 optical effect Effects 0.000 description 3
• 230000008520 organization Effects 0.000 description 3
• 230000003252 repetitive effect Effects 0.000 description 3
• 239000004065 semiconductor Substances 0.000 description 3
• RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 description 2
• 241000282326 Felis catus Species 0.000 description 2
• 238000003491 array Methods 0.000 description 2
• 238000013528 artificial neural network Methods 0.000 description 2
• 230000003190 augmentative effect Effects 0.000 description 2
• 229910052802 copper Inorganic materials 0.000 description 2
• 239000010949 copper Substances 0.000 description 2
• 230000001419 dependent effect Effects 0.000 description 2
• 238000009826 distribution Methods 0.000 description 2
• 239000000835 fiber Substances 0.000 description 2
• 230000006872 improvement Effects 0.000 description 2
• 230000006855 networking Effects 0.000 description 2
• 238000007781 pre-processing Methods 0.000 description 2
• 230000001902 propagating effect Effects 0.000 description 2
• 230000001131 transforming effect Effects 0.000 description 2
• 238000012384 transportation and delivery Methods 0.000 description 2
• 230000006978 adaptation Effects 0.000 description 1
• 230000003466 anti-cipated effect Effects 0.000 description 1
• 230000002457 bidirectional effect Effects 0.000 description 1
• 230000009172 bursting Effects 0.000 description 1
• 230000001413 cellular effect Effects 0.000 description 1
• 230000008878 coupling Effects 0.000 description 1
• 238000010168 coupling process Methods 0.000 description 1
• 238000005859 coupling reaction Methods 0.000 description 1
• 238000012517 data analytics Methods 0.000 description 1
• 238000013144 data compression Methods 0.000 description 1
• 238000013500 data storage Methods 0.000 description 1
• 238000013136 deep learning model Methods 0.000 description 1
• 230000002708 enhancing effect Effects 0.000 description 1
• 230000036541 health Effects 0.000 description 1
• 230000010354 integration Effects 0.000 description 1
• 238000002372 labelling Methods 0.000 description 1
• 239000002346 layers by function Substances 0.000 description 1
• 238000004519 manufacturing process Methods 0.000 description 1
• 238000013507 mapping Methods 0.000 description 1
• 230000007246 mechanism Effects 0.000 description 1
• 230000001537 neural effect Effects 0.000 description 1
• 239000013307 optical fiber Substances 0.000 description 1
• 238000013439 planning Methods 0.000 description 1
• 229920001690 polydopamine Polymers 0.000 description 1
• 238000011176 pooling Methods 0.000 description 1
• 238000013468 resource allocation Methods 0.000 description 1
• 230000003068 static effect Effects 0.000 description 1
• 238000007619 statistical method Methods 0.000 description 1
• 238000013526 transfer learning Methods 0.000 description 1
• 238000013519 translation Methods 0.000 description 1
• 230000014616 translation Effects 0.000 description 1
• 238000012795 verification Methods 0.000 description 1
• 238000012038 vulnerability analysis Methods 0.000 description 1

Images

Classifications

• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F8/00—Arrangements for software engineering
  • G06F8/30—Creation or generation of source code
  • G06F8/31—Programming languages or programming paradigms
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N3/00—Computing arrangements based on biological models
  • G06N3/02—Neural networks
  • G06N3/08—Learning methods
  • G06N3/088—Non-supervised learning, e.g. competitive learning
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F40/00—Handling natural language data
  • G06F40/40—Processing or translation of natural language
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F8/00—Arrangements for software engineering
  • G06F8/40—Transformation of program code
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06F—ELECTRIC DIGITAL DATA PROCESSING
  • G06F9/00—Arrangements for program control, e.g. control units
  • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
  • G06F9/44—Arrangements for executing specific programs
  • G06F9/445—Program loading or initiating
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N20/00—Machine learning
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N3/00—Computing arrangements based on biological models
  • G06N3/02—Neural networks
  • G06N3/04—Architecture, e.g. interconnection topology
  • G06N3/044—Recurrent networks, e.g. Hopfield networks
• • G—PHYSICS
  • G06—COMPUTING; CALCULATING OR COUNTING
  • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
  • G06N3/00—Computing arrangements based on biological models
  • G06N3/02—Neural networks
  • G06N3/04—Architecture, e.g. interconnection topology
  • G06N3/045—Combinations of networks

Definitions

• the present invention relates generally to the field of computing, and more particularly to natural language processing (NLP).
• NLP natural language processing
• Computer code is written in a style that is more repetitive than natural language. Natural language also has a limited number of words that can be used, however, computer programming code allows software developers to create new vocabulary or identifiers that may or may not contain useful information.
• Embodiments of the present invention disclose a method, a computer system, and a computer program product for building pre-trained contextual embeddings.
• Embodiments of the present invention may include collecting programming code.
• Embodiments of the present invention may include loading and preparing the programming code using a specialized programming language keywords-based vocabulary.
• Embodiments of the present invention may include creating contextual embeddings for the programming code.
• Embodiments of the present invention may include storing the contextual embeddings.
• FIG. 1 illustrates a networked computer environment according to at least one embodiment
• FIG. 2 is an operational flowchart illustrating a process for building pre-trained contextual embeddings for source code using specialized vocabulary according to at least one embodiment
• FIG. 3 is a block diagram of internal and external components of computers and servers depicted in FIG. 1 according to at least one embodiment
• FIG. 4 is a block diagram of an illustrative cloud computing environment including the computer system depicted in FIG. 1 , in accordance with an embodiment of the present disclosure.
• FIG. 5 is a block diagram of functional layers of the illustrative cloud computing environment of FIG. 5 , in accordance with an embodiment of the present disclosure.
• Unique concept keywords are words that may only be used in one context and may only have one specific meaning. Natural human language or domain adaptation does not have this unique concept of keywords. Many natural language words may have multiple meanings and may only be understood in a context of a whole phrase or sentence. Domain adaption, as it relates to machine learning or transfer learning, is a process that adapts a model from a source distribution to a target distribution.
• NLP natural language processing
• ML machine learning
• NLP natural language processing
• ML machine learning
• Probabilistic models may require different designs, modeling and learning for natural language and for programming language. Therefore, it may be advantageous to, among other things, create a system that both trains contextual embeddings for machine learning and solves the out-of-vocabulary (OOV) problem.
• OOV out-of-vocabulary
• embodiments of the present invention have the capacity to improve the field of software analysis and machine learning by creating a system that pertains contextual embeddings that also creates a deterministic solution for the out-of-vocabulary (OOV) problem. Improvements to machine learning model training and a reduced time to train a model is created by building contextual embeddings and using programming language keywords. More specifically, software analysis will be improved by building pre-trained contextual embeddings for source code using specialized vocabulary which will also solve the out-of-vocabulary (OOV) problem. Additional benefits may include solving programming code or source code issues, such as code searches, auto documentation, bug identification, vulnerability identification, code-based QA and programming language translations.
• Training data may include computer program code, such as source code, from various domain applications, such as retail, social media content, business, technology, medical, academic, government, industrial, food chain, legal or automotive.
• Models may be trained and updated with pre-trained contextual embeddings using supervised, unsupervised and semi-supervised machine learning procedures.
• Supervised learning may use a labeled dataset or a labeled training set to build, train and update a model.
• Unsupervised learning may use all unlabeled data to train a model.
• Semi-supervised learning may use both labeled datasets and unlabeled datasets to train a model.
• natural language processing may be used to extract tokens in source code to identify source code word contexts.
• a token may include a word or a part of a word.
• the word “language” may have one or more tokens associated with the word, such as one token may represent “lang”, “gua” or just one character “l”.
• a term (e.g., a word) or a token may refer to multiple potential meanings in natural language and in programming language, however, programming language may contain language specific keywords and the language specific keywords may only contain one meaning.
• natural language processing may also extract key features using word embedding techniques that map words and phrases to vectors of numbers using mathematical embeddings.
• Word embeddings may create an understanding of the occurrence of words or terms in a document corpus.
• the natural language processing accuracy may improve the semantic analysis.
• Pre-trained models and deep learning models that use natural language processing and word embeddings may provide more accurate predictive results.
• a pre-trained language model may include a model that is designed as a benchmark model for a particular dataset.
• a pre-trained model may also be called a natural language processing (NLP) model.
• the particular dataset used for pre-trained models may use contextual embeddings for learning.
• Contextual embeddings may typically be learned using natural language processing, however, since programming language may differ from natural language, pre-trained contextual embeddings may be created to for a pre-trained model to use for learning.
• Model training using natural language processing may be accomplished by feeding a model an unannotated dataset for training purposes. The unannotated or unstructured data may be classified or tagged for model training.
• Structured and unstructured data may be used as datasets to train contextual embeddings.
• Structured data may include data that is highly organized, such as labeled data (i.e., text file with labels), a spreadsheet or data that is stored in a fixed field.
• Unstructured data may include data that is not organized and has an unconventional internal structure, such as programming language, source code (e.g., a text file), a code snippet, a presentation, a webpage, an email, a word processing document or multimedia content.
• Computer program data may be used for the pre-training of contextual embeddings.
• the computer program data may include programming language documents such as source code data and application code data.
• source code data may be captured from an open source code repository or a private repository with proper accessibility permissions.
• Data may also be synthetically generated by programs that generate data.
• Source code may include data obtained from the various domain application repositories, such as telecommunication data, medical data, financial data, legal data, legislative data, business data, transportation data, agriculture data or industrial data. For example, a project in python language in a health care domain repository.
• Source code may be one type of programming code language that is available in large amounts via open-source code or also known as big code.
• the availability of the large amount of source code may assist in being utilized for analysis for new data-driven techniques, such as natural language processing (NLP) and machine learning (ML).
• NLP natural language processing
• ML machine learning
• natural language processing may ingest, via unsupervised machine learning, a large amount of input data to learn contextual embeddings for natural language processing tasks associate with machine learning.
• Source code may include computer programming language to be executed by a computing device, for example, source code project C and Java® (JavaScript and all JavaScript-based trademarks and logos are trademarks or registered trademarks of Oracle Corporation and/or its affiliates) source code project on Github® (Github and all Github-based trademarks and logos are trademarks or registered trademarks of Github, Inc. and/or its affiliates).
• Programming language can be viewed as a natural language that can be modeled using contextual embeddings.
• a transformer is a type of neural network architecture may also utilize pre-trained contextual embeddings to train a neural network by transforming one sequence into another.
• the transformer may have the ability to process program data.
• the transformer may ingest a sequence of tokens and may transform the tokens into pre-trained contextual embeddings.
• the granularity of the input tokens may be adjusted based on the tokenization technique used. For example, tokens may include single characters, characters plus keywords or subwords plus a keyword.
• the contextual embeddings may be used to pre-train a model, for example, by being fed into a transformer architecture to handle a sequence to sequence process or problem.
• the transformer may transform one sequence into another, for example, receive embeddings as an input and provide probabilities as an output.
• Sequence-2-sequence (seq2seq) may transform, for example, a sequence of words in the source code to another sequence of words.
• the contextual embeddings that are created may be an output of the transforming architecture.
• a long-short-term-memory (LSTM) model or architecture may also be used to analyze sequence dependent data.
• the long-short-term-memory (LSTM) model may also be used to create contextual embeddings by ingesting tokens based on a tokenization technique and provide an output of contextual embeddings for every token ingested or received.
• Contextual embeddings for a word may be captured semantically to bring the word into a correct or proper context.
• Semantic analysis may be used to infer the complexity of interactions, such as the meaning and intent of the language, such as the programming language or the source code for a computer. Semantic analysis may consider syntactic structures at various levels to infer meaning to a user's words, phrases or sentences.
• language specific keywords from programming languages may be used to create a programming language keywords-based vocabulary that is specialized for creating a deterministic solution for the out-of-vocabulary (OOV) problem.
• OOV out-of-vocabulary
• the out-of-vocabulary problem may occur from created identifiers, for example, by software developers that may include a created word that is not contained in a regular vocabulary for natural language.
• Identifiers may or may not contain useful information.
• an identifier may be created and called createNewVocabulary by a software developer.
• the identifier may be replaced by a tag, label or classifier during a preprocessing phase of machine learning. Replacing the identifier with a tag may lead to a loss of information since the identifier was a word or a phrase created by a developer for computer program code. Since the word or phrase may be a new word or phrase, the tag attached at the preprocessing phase may not be accurate.
• program language may also be writing in different styles.
• domain in a programming language may be considered different projects, such as FFmpeg or OpenSSL and each domain may have a different style of coding.
• Deterministic solutions provided herein to create efficient programming language modeling may include using a single character or a programming language-based vocabulary, by enhancing sub-word or byte pair encoding (BPE) vocabulary or by reducing the size of a large vocabulary to a smaller subset.
• BPE sub-word or byte pair encoding
• the deterministic solutions may be accomplished individually, in a combination or all together.
• contextual embeddings may be built and pre-trained by using tokens or tokenization techniques on programming code, such as source code, to feed as an input to a transformer or a long-short-term memory (LSTM) model.
• the output may include a vector for every input token.
• the vectors may represent the created contextual embeddings.
• the process of building or creating the vectors may be called training or pre-training.
• a single character-based vocabulary or a programming language-based vocabulary may be created using multiple methods.
• One method for creating a single character-based vocabulary may include tokenizing, for example, digits 0-9, lowercase a-z, uppercase A-Z or punctuation symbols.
• language specific keywords may be tokenized.
• Other vocabularies may be created by extending the character based vocabulary to a character plus keyword vocabulary by adding language specific keywords.
• subword tokens or byte pair encoding (BPE) vocabulary may be used and enhanced by adding language specific keywords to the vocabulary.
• language specific keywords may be added manually to the vocabulary for characters and the vocabulary for characters plus keywords.
• a subword-based vocabulary may be manually added if the subwords are not already present in the vocabulary.
• tokenization of a sentencepiece or a subword with a vocabulary size of 100 types may have 32 missing keywords.
• a type may include a character type, such as a letter, a number or a symbol.
• a type may also include a keyword type, such as part of a word or a whole word.
• Another example for a C language program may include a vocabulary size of 5000 with 1 missing keyword.
• program language modeling may be improved by using tokenization.
• tokenization For example, byte pair encoding (BPE) tokenization and vocabulary.
• BPE byte pair encoding
• approaches also known as sub-word or word piece approaches
• OOV out-of-vocabulary
• Byte pair encoding is a data compression scheme that requires a table of data to re-create the original data that was compressed based on common pairs of consecutive bytes.
• Augmenting byte pair encoding (BPE) vocabulary with programming language keywords may improve program language modeling. Augmenting the byte pair encoding (BPE) vocabulary may improve a model by ensuring that the keywords are not broken up by byte pair encoding (BPE), thus, the model may learn the embeddings more effectively and accurately.
• Byte pair encoding (BPE) vocabulary creation may be non-deterministic and may or may not contain programming language keywords. Sometimes the vocabulary creation may be a single character. Incorporating programming language keywords will create a deterministic way to augment the byte pair encoding (BPE) vocabulary and improve model performance.
• Byte pair encoding (BPE)-like vocabulary may also be reduced to create a faster model training time.
• Byte pair encoding (BPE)-like vocabularies can be arbitrarily large, and as such, model training time may take a longer amount of time to train using large vocabularies.
• Reducing the size of a large vocabulary to a smaller subset that may include keywords or characters can be accomplished by deleting some tokens.
• the vocabulary size may be reduced by deleting tokens that are not tokens based on the programming language keywords.
• large vocabulary may be reduced to a smaller subset by deleting types that are not related to keywords and characters.
• the vocabulary may contain a character type, such as “a” or a keyword type, such as “int” or any other word, such as “cat”.
• a character plus keyword vocabulary may be obtained by deleting “cat” so that only characters, such as “a” and keywords, such as “int” are left. For example, all tokens unrelated to keywords and characters in a vocabulary may be deleted.
• vocabulary may be created for machine learning training and model building.
• the vocabulary may be created manually, for example by adding subwords that are not already present in a vocabulary.
• the vocabulary may also be created automatically, for example, by using a byte pair encoding (BPE) library like sentencepiece.
• BPE byte pair encoding
• the vocabulary may be created to include a collection of type of words that are or that will be considered valid. For example, if the vocabulary contains “a” and “a” is a valid character type, then the tokenizer may recognize “a” when reading the data.
• a character tokenizer may have a character vocabulary, for example, “a”, “b”, “1”, “2”, “!”, etc.
• a small vocabulary may be created to both speed up programming language modeling and to solve the out-of-vocabulary (OOV) problem.
• the created vocabulary may include character level vocabulary with programming language keywords.
• a small vocabulary with only language specific keywords and single characters may be created.
• the character level vocabulary may be generalized for all programming languages in various domains that were created by many different software developers.
• the networked computer environment 100 may include a computer 102 with a processor 104 and a data storage device 106 that is enabled to run a software program 108 and a pre-trained embedding program 110 a .
• the networked computer environment 100 may also include a server 112 that is enabled to run a pre-trained embedding program 110 b that may interact with a database 114 and a communication network 116 .
• the networked computer environment 100 may include a plurality of computers 102 and servers 112 , only one of which is shown.
• the communication network 116 may include various types of communication networks, such as a wide area network (WAN), local area network (LAN), a telecommunication network, a wireless network, a public switched network and/or a satellite network.
• WAN wide area network
• LAN local area network
• the client computer 102 may communicate with the server computer 112 via the communications network 116 .
• the communications network 116 may include connections, such as wire, wireless communication links, or fiber optic cables.
• server computer 112 may include internal components 902 a and external components 904 a , respectively, and client computer 102 may include internal components 902 b and external components 904 b , respectively.
• Server computer 112 may also operate in a cloud computing service model, such as Software as a Service (SaaS), Analytics as a Service (AaaS), Blockchain as a Service (BaaS), Platform as a Service (PaaS), or Infrastructure as a Service (IaaS).
• SaaS Software as a Service
• AaaS Analytics as a Service
• BaaS Blockchain as a Service
• PaaS Platform as a Service
• IaaS Infrastructure as a Service
• Server 112 may also be located in a cloud computing deployment model, such as a private cloud, community cloud, public cloud, or hybrid cloud.
• Client computer 102 may be, for example, a mobile device, a telephone, a personal digital assistant, a netbook, a laptop computer, a tablet computer, a desktop computer, or any type of computing devices capable of running a program, accessing a network, and accessing a database 114 .
• the pre-trained embedding program 110 a , 110 b may interact with a database 114 that may be embedded in various storage devices, such as, but not limited to a computer/mobile device 102 , a networked server 112 , or a cloud storage service.
• a user using a client computer 102 or a server computer 112 may use the pre-trained embedding program 110 a , 110 b (respectively) to analyze, detect and predict vulnerabilities of a computing system using contextual embeddings.
• the contextual embedding method is explained in more detail below with respect to FIG. 2 .
• FIG. 2 an operational flowchart illustrating the exemplary building pre-trained contextual embeddings for source code process 200 used by the pre-trained embedding program 110 a , 110 b according to at least one embodiment is depicted.
• Programming code is collected to build pre-trained data.
• Programming code may include, for example, source code collected and used for the purposes of training a model for a pre-training phase.
• the source code may be collected from an open source computer programming repository.
• the source code may be used as training data to create and build contextual embeddings for programming language.
• Source code may be collected from various repositories that store source code for multiple domains.
• Source code collected for training may be obtained, for example, from source code projects or an open source accessibility to source code.
• Source code and other types of computer programming code may be collected from domain specific applications with proper accessibility to the source code repositories.
• programming code data is loaded and prepared.
• the collected programming code such as source code
• Data loading may include reorganizing the training data to a form used for training if the data obtained from the repository is not already in the required form.
• Raw data that is collected may often contain comments or empty lines that may not be useful and may be removed during data loading.
• the data loader can reorganize or transform the source file data to information that the model can use or consume for training.
• Loading and preparing data may use programming language keywords to create specialized tokenizers.
• the loading and preparing the programming code may use a specialized programming language keywords-based vocabulary.
• the training data may be tokenized. Tokenizing the training data may be accomplished at a character level or at a single character plus keyword level. For example, a character tokenizer may be used at a character level. Character tokenization may be extended to include programming language keywords as types. Tokenizing the training data may also be accomplished by adding keywords as types. Tokenizing training data may also be accomplished by using a sub-word method, such as byte pair encoding (BPE).
• BPE byte pair encoding
• Subword techniques may also be extended by including programming language keywords as types.
• the process of vocabulary creation using subword tokenization may or may not include keywords, however, if keywords are present, they may be included.
• the contextual embeddings are pre-trained for programming code.
• the process of creating contextual embeddings may also be known as pre-training or training.
• contextual embeddings may be created as pre-training data and the embeddings may used for fine tuning.
• the pre-training may consist of unsupervised learning on, for example, transformer architecture, natural language models or machine learning models to apply training to the contextual embeddings.
• the collected, loaded and prepared programming code or source code with the built contextual embeddings may be considered model training data and allocated for model pre-training or model training.
• the model training may include, for example, utilizing the training data as data to train a natural language model or a machine learning model.
• the training data may include the ability to use a smaller dataset to train the model for a specific task.
• the natural language model may be trained, for example, as a pre-trained model using a smaller dataset, such as source code, to perform a specific task, such as creating embeddings for the source code.
• Other pre-trained models may also be used with the collected training data, such as bidirectional encoders to encode the loaded training data.
• the pre-trained contextual embeddings are stored.
• the pre-trained contextual embeddings that have been built and created for programming code or source code may be stored in a file to be used in conjunction with other files, such as configuration files, vocabulary files and other files depending on the usage of the files.
• the pre-trained contextual embeddings may be stored in a repository or a database for later use.
• the database that stores the trained model may be accessed during use or future use, such as for pre-train model building, a fine-tuning phase or other machine learning uses.
• the pre-trained contextual embeddings have learned how programming language is written and as such, can perform many downstream tasks that may be built and used in addition to the pre-trained model. Other downstream tasks may include a fine-tuning phase, added classification and labels of datasets to perform particular machine learning tasks or to use the pre-trained model as is or as a machine learning model.
• FIG. 2 provides only an illustration of one embodiment and do not imply any limitations with regard to how different embodiments may be implemented. Many modifications to the depicted embodiment(s) may be made based on design and implementation requirements.
• FIG. 3 is a block diagram 900 of internal and external components of computers depicted in FIG. 1 in accordance with an illustrative embodiment of the present invention. It should be appreciated that FIG. 3 provides only an illustration of one implementation and does not imply any limitations with regard to the environments in which different embodiments may be implemented. Many modifications to the depicted environments may be made based on design and implementation requirements.
• Data processing system 902 , 904 is representative of any electronic device capable of executing machine-readable program instructions.
• Data processing system 902 , 904 may be representative of a smart phone, a computer system, PDA, or other electronic devices.
• Examples of computing systems, environments, and/or configurations that may represented by data processing system 902 , 904 include, but are not limited to, personal computer systems, server computer systems, thin clients, thick clients, hand-held or laptop devices, multiprocessor systems, microprocessor-based systems, network PCs, minicomputer systems, and distributed cloud computing environments that include any of the above systems or devices.
• User client computer 102 and network server 112 may include respective sets of internal components 902 a, b and external components 904 a, b illustrated in FIG. 3 .
• Each of the sets of internal components 902 a, b includes one or more processors 906 , one or more computer-readable RAMs 908 and one or more computer-readable ROMs 910 on one or more buses 912 , and one or more operating systems 914 and one or more computer-readable tangible storage devices 916 .
• the one or more operating systems 914 , the software program 108 , and the pre-trained embedding program 110 a in client computer 102 , and the pre-trained embedding program 110 b in network server 112 may be stored on one or more computer-readable tangible storage devices 916 for execution by one or more processors 906 via one or more RAMs 908 (which typically include cache memory).
• each of the computer-readable tangible storage devices 916 is a magnetic disk storage device of an internal hard drive.
• each of the computer-readable tangible storage devices 916 is a semiconductor storage device such as ROM 910 , EPROM, flash memory or any other computer-readable tangible storage device that can store a computer program and digital information.
• Each set of internal components 902 a, b also includes a R/W drive or interface 918 to read from and write to one or more portable computer-readable tangible storage devices 920 such as a CD-ROM, DVD, memory stick, magnetic tape, magnetic disk, optical disk or semiconductor storage device.
• a software program such as the software program 108 and the pre-trained embedding program 110 a , 110 b can be stored on one or more of the respective portable computer-readable tangible storage devices 920 , read via the respective R/W drive or interface 918 and loaded into the respective hard drive 916 .
• Each set of internal components 902 a, b may also include network adapters (or switch port cards) or interfaces 922 such as a TCP/IP adapter cards, wireless wi-fi interface cards, or 3G or 4G wireless interface cards or other wired or wireless communication links.
• the software program 108 and the pre-trained embedding program 110 a in client computer 102 and the pre-trained embedding program 110 b in network server computer 112 can be downloaded from an external computer (e.g., server) via a network (for example, the Internet, a local area network or other, wide area network) and respective network adapters or interfaces 922 .
• the network may comprise copper wires, optical fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers.
• Each of the sets of external components 904 a, b can include a computer display monitor 924 , a keyboard 926 , and a computer mouse 928 .
• External components 904 a, b can also include touch screens, virtual keyboards, touch pads, pointing devices, and other human interface devices.
• Each of the sets of internal components 902 a, b also includes device drivers 930 to interface to computer display monitor 924 , keyboard 926 and computer mouse 928 .
• the device drivers 930 , R/W drive or interface 918 and network adapter or interface 922 comprise hardware and software (stored in storage device 916 and/or ROM 910 ).
• Cloud computing is a model of service delivery for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g. networks, network bandwidth, servers, processing, memory, storage, applications, virtual machines, and services) that can be rapidly provisioned and released with minimal management effort or interaction with a provider of the service.
• This cloud model may include at least five characteristics, at least three service models, and at least four deployment models.
• On-demand self-service a cloud consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with the service's provider.
• Resource pooling the provider's computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to demand. There is a sense of location independence in that the consumer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter).
• Rapid elasticity capabilities can be rapidly and elastically provisioned, in some cases automatically, to quickly scale out and rapidly released to quickly scale in. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be purchased in any quantity at any time.
• Measured service cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported providing transparency for both the provider and consumer of the utilized service.
• level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts).
• SaaS Software as a Service: the capability provided to the consumer is to use the provider's applications running on a cloud infrastructure or on a hybrid cloud infrastructure.
• the applications are accessible from various client devices through a thin client interface such as a web browser (e.g., web-based e-mail).
• a web browser e.g., web-based e-mail
• the consumer does not manage or control the underlying cloud infrastructure including network, servers, operating systems, storage, or even individual application capabilities, with the possible exception of limited user-specific application configuration settings.
• PaaS Platform as a Service
• the consumer does not manage or control the underlying cloud infrastructure including networks, servers, operating systems, or storage, but has control over the deployed applications and possibly application hosting environment configurations.
• Analytics as a Service the capability provided to the consumer is to use web-based or cloud-based networks (i.e., infrastructure) to access an analytics platform.
• Analytics platforms may include access to analytics software resources or may include access to relevant databases, corpora, servers, operating systems or storage.
• the consumer does not manage or control the underlying web-based or cloud-based infrastructure including databases, corpora, servers, operating systems or storage, but has control over the deployed applications and possibly application hosting environment configurations.
• IaaS Infrastructure as a Service
• the consumer does not manage or control the underlying cloud infrastructure but has control over operating systems, storage, deployed applications, and possibly limited control of select networking components (e.g., host firewalls).
• Private cloud the cloud infrastructure is operated solely for an organization. It may be managed by the organization or a third party and may exist on-premises or off-premises.
• Public cloud the cloud infrastructure is made available to the general public or a large industry group and is owned by an organization selling cloud services.
• Hybrid cloud the cloud infrastructure is a composition of two or more clouds (private, community, or public) that remain unique entities but are bound together by standardized or proprietary technology that enables data and application portability (e.g., cloud bursting for load-balancing between clouds).
• a cloud computing environment is service oriented with a focus on statelessness, low coupling, modularity, and semantic interoperability.
• An infrastructure comprising a network of interconnected nodes.
• cloud computing environment 1000 comprises one or more cloud computing nodes 100 with which local computing devices used by cloud consumers, such as, for example, personal digital assistant (PDA) or cellular telephone 1000 A, desktop computer 1000 B, laptop computer 1000 C, and/or automobile computer system 1000 N may communicate.
• Nodes 100 may communicate with one another. They may be grouped (not shown) physically or virtually, in one or more networks, such as Private, Community, Public, or Hybrid clouds as described hereinabove, or a combination thereof.
• This allows cloud computing environment 1000 to offer infrastructure, platforms and/or software as services for which a cloud consumer does not need to maintain resources on a local computing device.
• computing devices 1000 A-N shown in FIG. 4 are intended to be illustrative only and that computing nodes 100 and cloud computing environment 1000 can communicate with any type of computerized device over any type of network and/or network addressable connection (e.g., using a web browser).
• FIG. 5 a set of functional abstraction layers 1100 provided by cloud computing environment 1000 is shown. It should be understood in advance that the components, layers, and functions shown in FIG. 5 are intended to be illustrative only and embodiments of the invention are not limited thereto. As depicted, the following layers and corresponding functions are provided:
• Hardware and software layer 1102 includes hardware and software components.
• hardware components include: mainframes 1104 ; RISC (Reduced Instruction Set Computer) architecture based servers 1106 ; servers 1108 ; blade servers 1110 ; storage devices 1112 ; and networks and networking components 1114 .
• software components include network application server software 1116 and database software 1118 .
• Virtualization layer 1120 provides an abstraction layer from which the following examples of virtual entities may be provided: virtual servers 1122 ; virtual storage 1124 ; virtual networks 1126 , including virtual private networks; virtual applications and operating systems 1128 ; and virtual clients 1130 .
• management layer 1132 may provide the functions described below.
• Resource provisioning 1134 provides dynamic procurement of computing resources and other resources that are utilized to perform tasks within the cloud computing environment.
• Metering and Pricing 1136 provide cost tracking as resources are utilized within the cloud computing environment, and billing or invoicing for consumption of these resources. In one example, these resources may comprise application software licenses.
• Security provides identity verification for cloud consumers and tasks, as well as protection for data and other resources.
• User portal 1138 provides access to the cloud computing environment for consumers and system administrators.
• Service level management 1140 provides cloud computing resource allocation and management such that required service levels are met.
• Service Level Agreement (SLA) planning and fulfillment 1142 provide pre-arrangement for, and procurement of, cloud computing resources for which a future requirement is anticipated in accordance with an SLA.
• SLA Service Level Agreement
• Workloads layer 1144 provides examples of functionality for which the cloud computing environment may be utilized. Examples of workloads and functions which may be provided from this layer include: mapping and navigation 1146 ; software development and lifecycle management 1148 ; virtual classroom education delivery 1150 ; data analytics processing 1152 ; transaction processing 1154 ; and vulnerability analysis using contextual embeddings 1156 .
• a pre-trained embedding program 110 a , 110 b provides a way to build pre-trained contextual embeddings for source code using specialized vocabulary.
• the present invention may be a system, a method, and/or a computer program product at any possible technical detail level of integration
• the computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention
• the computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device.
• the computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing.
• a non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing.
• RAM random access memory
• ROM read-only memory
• EPROM or Flash memory erasable programmable read-only memory
• SRAM static random access memory
• CD-ROM compact disc read-only memory
• DVD digital versatile disk
• memory stick a floppy disk
• a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon
• a computer readable storage medium is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.
• Computer readable program instructions described herein can be downloaded to respective computing/processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and/or a wireless network.
• the network may comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers.
• a network adapter card or network interface in each computing/processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing/processing device.
• Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction-set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the “C” programming language, python programming language or similar programming languages.
• the computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server.
• the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider).
• electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), or programmable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.
• These computer readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks.
• These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, a programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable storage medium having instructions stored therein comprises an article of manufacture including instructions which implement aspects of the function/act specified in the flowchart and/or block diagram block or blocks.
• the computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other device to produce a computer implemented process, such that the instructions which execute on the computer, other programmable apparatus, or other device implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
• each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s).
• the functions noted in the blocks may occur out of the order noted in the Figures.
• two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved.

Landscapes

• Engineering & Computer Science (AREA)
• Theoretical Computer Science (AREA)
• Software Systems (AREA)
• General Engineering & Computer Science (AREA)
• Physics & Mathematics (AREA)
• General Physics & Mathematics (AREA)
• Computing Systems (AREA)
• Artificial Intelligence (AREA)
• Mathematical Physics (AREA)
• Health & Medical Sciences (AREA)
• General Health & Medical Sciences (AREA)
• Evolutionary Computation (AREA)
• Data Mining & Analysis (AREA)
• Computational Linguistics (AREA)
• Computer Vision & Pattern Recognition (AREA)
• Medical Informatics (AREA)
• Audiology, Speech & Language Pathology (AREA)
• Biomedical Technology (AREA)
• Life Sciences & Earth Sciences (AREA)
• Biophysics (AREA)
• Molecular Biology (AREA)
• Stored Programmes (AREA)

Abstract

Description

Claims (17)

US16/917,967 2020-07-01 2020-07-01 Building pre-trained contextual embeddings for programming languages using specialized vocabulary Active 2040-10-21 US11429352B2 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title

US16/917,967 US11429352B2 (en)	2020-07-01	2020-07-01	Building pre-trained contextual embeddings for programming languages using specialized vocabulary

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title

US16/917,967 US11429352B2 (en)	2020-07-01	2020-07-01	Building pre-trained contextual embeddings for programming languages using specialized vocabulary

Publications (2)

ID=79166918

Family Applications (1)

Application Number Title Priority Date Filing Date

US16/917,967 Active 2040-10-21 US11429352B2 (en)	2020-07-01	2020-07-01	Building pre-trained contextual embeddings for programming languages using specialized vocabulary

Country Status (1)

Cited By (2)

* Cited by examiner, † Cited by third party Publication number Priority date Publication date Assignee Title

US20250130780A1 (en) *	2023-10-19	2025-04-24	Oracle International Corporation	Developing a programming language model for machine learning tasks
US12307247B1 (en)	2024-02-02	2025-05-20	Marek Dziubinski	Transformer-based programming code value quantification system

Families Citing this family (6)

* Cited by examiner, † Cited by third party Publication number Priority date Publication date Assignee Title

US11893363B2 (en) *	2020-08-27	2024-02-06	Microsoft Technology Licensing, Llc.	Unit test case generation with transformers
US11651014B2 (en) *	2020-10-30	2023-05-16	Fujitsu Limited	Source code retrieval
US11797281B2 (en) *	2021-08-05	2023-10-24	Sap Se	Multi-language source code search engine
CN114968412B (en) *	2022-06-20	2024-02-02	中国平安财产保险股份有限公司	Configuration file generation method, device, equipment and medium based on artificial intelligence
US12141553B2 (en)	2022-06-22	2024-11-12	Amazon Technologies, Inc.	Programmatically generating evaluation data sets for code generation models
US20230419036A1 (en) *	2022-06-22	2023-12-28	Amazon Technologies, Inc.	Random token segmentation for training next token prediction models

Citations (23)

* Cited by examiner, † Cited by third party Publication number Priority date Publication date Assignee Title

US20160212153A1 (en)	2015-01-16	2016-07-21	Microsoft Technology Licensing, Llc.	Code Labeling Based on Tokenized Code Samples
US9935972B2 (en)	2015-06-29	2018-04-03	Fortinet, Inc.	Emulator-based malware learning and detection
US20180157644A1 (en)	2016-12-07	2018-06-07	Disney Enterprises, Inc.	Dynamic word embeddings
CN108549817A (en)	2018-04-19	2018-09-18	北京理工大学	A kind of software security flaw prediction technique based on text deep learning
US20180285397A1 (en)	2017-04-04	2018-10-04	Cisco Technology, Inc.	Entity-centric log indexing with context embedding
US20190050319A1 (en)	2017-08-08	2019-02-14	Accenture Global Solutions Limited	Intellectual automated security, performance and code generation framework
US20190197097A1 (en) *	2017-12-22	2019-06-27	International Business Machines Corporation	Cognitive framework to detect adverse events in free-form text
US20190215329A1 (en)	2018-01-08	2019-07-11	Sophos Limited	Malware detection using machine learning
US20190324744A1 (en)	2019-06-28	2019-10-24	Intel Corporation	Methods, systems, articles of manufacture, and apparatus for a context and complexity-aware recommendation system for improved software development efficiency
US20190325273A1 (en)	2018-04-18	2019-10-24	Sony Interactive Entertainment Inc.	Context embedding for capturing image dynamics
US20190370473A1 (en)	2018-05-30	2019-12-05	Nvidia Corporation	Detecting vulnerabilities to fault injection in computer code using machine learning
CN110688861A (en)	2019-09-26	2020-01-14	沈阳航空航天大学	Multi-feature fusion sentence-level translation quality estimation method
US20200050207A1 (en)	2018-08-09	2020-02-13	GM Global Technology Operations LLC	Systems, apparatus, and methods for embedded encodings of contextual information using a neural network with vector space modeling
US10581888B1 (en)	2017-07-31	2020-03-03	EMC IP Holding Company LLC	Classifying software scripts utilizing deep learning networks
US20200076841A1 (en)	2018-09-05	2020-03-05	Oracle International Corporation	Context-aware feature embedding and anomaly detection of sequential log data using deep recurrent neural networks
US20200076842A1 (en)	2018-09-05	2020-03-05	Oracle International Corporation	Malicious activity detection by cross-trace analysis and deep learning
US20200097809A1 (en)	2018-09-24	2020-03-26	Salesforce.Com, Inc.	Case Object Context Embeddings for Machine Learning Training of Case Context
US10606946B2 (en)	2015-07-06	2020-03-31	Microsoft Technology Licensing, Llc	Learning word embedding using morphological knowledge
US20200183681A1 (en) *	2018-12-11	2020-06-11	Sap Se	Method for a software development system
US20200249918A1 (en) *	2019-02-02	2020-08-06	Microsoft Technology Licensing, Llc.	Deep learning enhanced code completion system
US20210056211A1 (en)	2019-08-23	2021-02-25	Praetorian	System and method for automatically detecting a security vulnerability in a source code using a machine learning model
US20210327413A1 (en) *	2020-04-16	2021-10-21	Microsoft Technology Licensing, Llc	Natural language processing models for conversational computing
US20220004642A1 (en)	2020-07-01	2022-01-06	International Business Machines Corporation	Vulnerability analysis using contextual embeddings

• 2020
  • 2020-07-01 US US16/917,967 patent/US11429352B2/en active Active

Patent Citations (25)

* Cited by examiner, † Cited by third party Publication number Priority date Publication date Assignee Title

US20160212153A1 (en)	2015-01-16	2016-07-21	Microsoft Technology Licensing, Llc.	Code Labeling Based on Tokenized Code Samples
US9935972B2 (en)	2015-06-29	2018-04-03	Fortinet, Inc.	Emulator-based malware learning and detection
US10606946B2 (en)	2015-07-06	2020-03-31	Microsoft Technology Licensing, Llc	Learning word embedding using morphological knowledge
US20180157644A1 (en)	2016-12-07	2018-06-07	Disney Enterprises, Inc.	Dynamic word embeddings
US20180285397A1 (en)	2017-04-04	2018-10-04	Cisco Technology, Inc.	Entity-centric log indexing with context embedding
US10581888B1 (en)	2017-07-31	2020-03-03	EMC IP Holding Company LLC	Classifying software scripts utilizing deep learning networks
US20190050319A1 (en)	2017-08-08	2019-02-14	Accenture Global Solutions Limited	Intellectual automated security, performance and code generation framework
US20190197097A1 (en) *	2017-12-22	2019-06-27	International Business Machines Corporation	Cognitive framework to detect adverse events in free-form text
US10540438B2 (en) *	2017-12-22	2020-01-21	International Business Machines Corporation	Cognitive framework to detect adverse events in free-form text
US20190215329A1 (en)	2018-01-08	2019-07-11	Sophos Limited	Malware detection using machine learning
US20190325273A1 (en)	2018-04-18	2019-10-24	Sony Interactive Entertainment Inc.	Context embedding for capturing image dynamics
CN108549817A (en)	2018-04-19	2018-09-18	北京理工大学	A kind of software security flaw prediction technique based on text deep learning
US20190370473A1 (en)	2018-05-30	2019-12-05	Nvidia Corporation	Detecting vulnerabilities to fault injection in computer code using machine learning
US20200050207A1 (en)	2018-08-09	2020-02-13	GM Global Technology Operations LLC	Systems, apparatus, and methods for embedded encodings of contextual information using a neural network with vector space modeling
US20200076842A1 (en)	2018-09-05	2020-03-05	Oracle International Corporation	Malicious activity detection by cross-trace analysis and deep learning
US20200076841A1 (en)	2018-09-05	2020-03-05	Oracle International Corporation	Context-aware feature embedding and anomaly detection of sequential log data using deep recurrent neural networks
US20200097809A1 (en)	2018-09-24	2020-03-26	Salesforce.Com, Inc.	Case Object Context Embeddings for Machine Learning Training of Case Context
US20200183681A1 (en) *	2018-12-11	2020-06-11	Sap Se	Method for a software development system
US20200249918A1 (en) *	2019-02-02	2020-08-06	Microsoft Technology Licensing, Llc.	Deep learning enhanced code completion system
US20210271455A1 (en) *	2019-02-02	2021-09-02	Microsoft Technology Licensing, Llc.	Deep learning enhanced code completion system
US20190324744A1 (en)	2019-06-28	2019-10-24	Intel Corporation	Methods, systems, articles of manufacture, and apparatus for a context and complexity-aware recommendation system for improved software development efficiency
US20210056211A1 (en)	2019-08-23	2021-02-25	Praetorian	System and method for automatically detecting a security vulnerability in a source code using a machine learning model
CN110688861A (en)	2019-09-26	2020-01-14	沈阳航空航天大学	Multi-feature fusion sentence-level translation quality estimation method
US20210327413A1 (en) *	2020-04-16	2021-10-21	Microsoft Technology Licensing, Llc	Natural language processing models for conversational computing
US20220004642A1 (en)	2020-07-01	2022-01-06	International Business Machines Corporation	Vulnerability analysis using contextual embeddings

Non-Patent Citations (25)

* Cited by examiner, † Cited by third party Title

Allamanis et al., "A Survey of Machine Learning for Big Code and Naturalness", May 5, 2018, pp. 1-36.
Allamanis et al., "Learning To Represent Programs With Graphs", published as a conference paper at ICLR 2018, May 4, 2018, pp. 1-17.
Anonymous authors, "Pre-Trained Contextual Embedding of Source Code", [Online], pp. 1-20, [Retrieved from Internet on Jun. 18, 2022], <https://openreview.net/pdf?id=rygoURNYvS>, (Year: 2019). *
Babii et al., "Modeling Vocabulary for Big Code Machine Learning", Apr. 3, 2019, pp. 1-12.
Beltagy et al., "SciBERT: A Pretrained Language Model for Scientific Text", Sep. 10, 2019, 6 pages.
Chen et al., "A Literature Study of Embeddings on Source Code", Apr. 5, 2019, pp. 1-8.
Chen et al., "Strategies for Training Large Vocabulary Neural Language Models", Dec. 15, 2015, 13 pages.
Devlin et al., "BERT: Pre-training of Deep BiDirectional Transformers for Language Understanding", May 24, 2019, 16 pages.
Feng et al., "CodeBERT: A pre-Trained Model for Programming and Natural Languages", Feb. 19, 2020, 11 pages.
Grace Period Disclosure, Buratti et al., "Exploring Software Naturalness through Neural Language Models", arXiv:2006.12641V2 [cs.CL] Jun. 24, 2020, 12 pages.
Hindle et al., "On the Naturalness of Software", ICSE 2012, Zurich, Switzerland, 2012 IEEE, pp. 837-847.
IBM: List of IBM Patents or Patent Applications Treated as Related (Appendix P), Jul. 6, 2020, 2 pages.
Jongho Shin et al., "Cross-domain meta-learning for bug finding in the source codes with a small dataset", [Online], pp. 1-6, [Retrieved from Internet on Jun. 18, 2022], <https://dl.acm.org/doi/pdf/10.1145/3424954.3424957>, (Year: 2020). *
Kanade et al., "Pre-trained Contextual Embedding of Source Code", Jan. 3, 2020, pp. 1-22.
Karampatsis et al., "Maybe Deep Neural Networks are the Best Choice for Modeling Source Code", Mar. 13, 2019, 12 pages.
Lee et al., "BioBERT: a pre-trained biomedical language representation model for biomedical text mining", Bioinformatics, 2019, 1-7, Advance Access Publication Date: Sep. 10, 2019, Original paper, OXFORD.
Li et al., "VulDeepecker: A Deep Learning-Based System for Vulnerability Detection", Network and Distributed Systems Security (NDSS) Symposium 2018,Feb. 18-21, 2018, San Diego, CA, USA, 15 pages.
Mell et al., "The NIST Definition of Cloud Computing", Recommendations of the National Institute of Standards and Technology, Special Publication 800-145, Sep. 2011, 7 pages.
Pujar et al., "Vulnerability Analysis Using Contextual Embeddings", U.S. Appl. No. 16/917,962, filed Jul. 1, 2020, 34 pages.
Ray et al., "On the ‘Naturalness’ of Buggy Code", 2016 IEEE/ACM 38th IEEE International Conference on Software Engineering, ICSE' 16, May 14-22, 2016, Austin, TX, USA, pp. 428-439.
Raychev et al., "Code Completion with Statistical Language Models", PLDI 2014, Jun. 9-11, 2014, Edinburgh, United Kingdom, 2014 ACM, pp. 419-428.
Russell et al., "Automated Vulnerability Detection in Source Code Using Deep Representation Learning", 17th IEEE International Conference on Machine Learning and Applications (IEEE ICMLA 2018), Orlando, Florida, USA, 7 pages.
Tai, "Effects of Inserting Domain Vocabulary and Fine-tuning BERT for German Legal Language", University of Twente, deepset, Master's Thesis, Nov. 26, 2019, 91 pages.
Vaswani et al., "Attention is all You Need", Dec. 6, 2017, 31st Conference on Neural Information processing Systems (NIPS 2017), Long Beach, CA, USA, 15 pages.
Zhou et al., "Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks", 33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada, 11 pages.

Cited By (2)

* Cited by examiner, † Cited by third party Publication number Priority date Publication date Assignee Title

US20250130780A1 (en) *	2023-10-19	2025-04-24	Oracle International Corporation	Developing a programming language model for machine learning tasks
US12307247B1 (en)	2024-02-02	2025-05-20	Marek Dziubinski	Transformer-based programming code value quantification system

Also Published As

Similar Documents

Legal Events

Date Code Title Description

2020-07-01	AS	Assignment	Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW YORK Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:PUJAR, SAURABH;BURATTI, LUCA;MORARI, ALESSANDRO;AND OTHERS;SIGNING DATES FROM 20200625 TO 20200626;REEL/FRAME:053096/0046
2020-07-01	FEPP	Fee payment procedure	Free format text: ENTITY STATUS SET TO UNDISCOUNTED (ORIGINAL EVENT CODE: BIG.); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY
2021-12-20	STPP	Information on status: patent application and granting procedure in general	Free format text: NON FINAL ACTION MAILED
2022-02-17	STPP	Information on status: patent application and granting procedure in general	Free format text: RESPONSE TO NON-FINAL OFFICE ACTION ENTERED AND FORWARDED TO EXAMINER
2022-03-29	STPP	Information on status: patent application and granting procedure in general	Free format text: FINAL REJECTION MAILED
2022-06-06	STPP	Information on status: patent application and granting procedure in general	Free format text: ADVISORY ACTION MAILED
2022-06-15	STPP	Information on status: patent application and granting procedure in general	Free format text: DOCKETED NEW CASE - READY FOR EXAMINATION
2022-06-24	STPP	Information on status: patent application and granting procedure in general	Free format text: NOTICE OF ALLOWANCE MAILED -- APPLICATION RECEIVED IN OFFICE OF PUBLICATIONS
2022-08-03	STPP	Information on status: patent application and granting procedure in general	Free format text: PUBLICATIONS -- ISSUE FEE PAYMENT VERIFIED
2022-08-10	STCF	Information on status: patent grant	Free format text: PATENTED CASE