JP7778160B2 - 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via) - Google Patents

単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via)

Info

Publication number
JP7778160B2
JP7778160B2 JP2023568351A JP2023568351A JP7778160B2 JP 7778160 B2 JP7778160 B2 JP 7778160B2 JP 2023568351 A JP2023568351 A JP 2023568351A JP 2023568351 A JP2023568351 A JP 2023568351A JP 7778160 B2 JP7778160 B2 JP 7778160B2
Authority
JP
Japan
Prior art keywords
samples
adversarial
machine learning
model
learning model
Prior art date
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
Application number
JP2023568351A
Other languages
English (en)
Japanese (ja)
Other versions
JP2024518416A5 (cg-RX-API-DMAC7.html
JP2024518416A (ja
Inventor
ガッデ,シュリニバーサ・ファニ・クマール
ジャオ,シュ
Original Assignee
オラクル・インターナショナル・コーポレイション
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.)
Filing date
Publication date
Application filed by オラクル・インターナショナル・コーポレイション filed Critical オラクル・インターナショナル・コーポレイション
Publication of JP2024518416A publication Critical patent/JP2024518416A/ja
Publication of JP2024518416A5 publication Critical patent/JP2024518416A5/ja
Priority to JP2025197460A priority Critical patent/JP2026035651A/ja
Application granted granted Critical
Publication of JP7778160B2 publication Critical patent/JP7778160B2/ja
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/60—Protecting data
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G06F40/295—Named entity recognition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Bioethics (AREA)
  • Computer Hardware Design (AREA)
  • Computer Security & Cryptography (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Artificial Intelligence (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Machine Translation (AREA)
JP2023568351A 2021-05-07 2022-03-29 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via) Active JP7778160B2 (ja)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2025197460A JP2026035651A (ja) 2021-05-07 2025-11-18 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via)

Applications Claiming Priority (3)

Application Number Priority Date Filing Date Title
US202163185531P 2021-05-07 2021-05-07
US63/185,531 2021-05-07
PCT/US2022/022291 WO2022235353A1 (en) 2021-05-07 2022-03-29 Variant inconsistency attack (via) as a simple and effective adversarial attack method

Related Child Applications (1)

Application Number Title Priority Date Filing Date
JP2025197460A Division JP2026035651A (ja) 2021-05-07 2025-11-18 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via)

Publications (3)

Publication Number Publication Date
JP2024518416A JP2024518416A (ja) 2024-05-01
JP2024518416A5 JP2024518416A5 (cg-RX-API-DMAC7.html) 2024-12-24
JP7778160B2 true JP7778160B2 (ja) 2025-12-01

Family

ID=81346304

Family Applications (2)

Application Number Title Priority Date Filing Date
JP2023568351A Active JP7778160B2 (ja) 2021-05-07 2022-03-29 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via)
JP2025197460A Pending JP2026035651A (ja) 2021-05-07 2025-11-18 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via)

Family Applications After (1)

Application Number Title Priority Date Filing Date
JP2025197460A Pending JP2026035651A (ja) 2021-05-07 2025-11-18 単純で効果的な敵対的攻撃方法としてのバリアント不一致攻撃(via)

Country Status (5)

Country Link
US (2) US12086274B2 (cg-RX-API-DMAC7.html)
EP (1) EP4334832A1 (cg-RX-API-DMAC7.html)
JP (2) JP7778160B2 (cg-RX-API-DMAC7.html)
CN (1) CN117296058A (cg-RX-API-DMAC7.html)
WO (1) WO2022235353A1 (cg-RX-API-DMAC7.html)

Families Citing this family (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11316806B1 (en) * 2020-01-28 2022-04-26 Snap Inc. Bulk message deletion
EP4334832A1 (en) 2021-05-07 2024-03-13 Oracle International Corporation Variant inconsistency attack (via) as a simple and effective adversarial attack method
US11776241B1 (en) * 2022-09-30 2023-10-03 Fudan University Method and system for generating adversarial attacks on video recognition models
US12585682B2 (en) * 2022-10-10 2026-03-24 Tata Consultancy Services Limited Method and system for generating longform technical question and answer dataset
US12701130B2 (en) * 2022-12-27 2026-08-04 Cognyte Technologies Israel Ltd Systems and methods for using machine learning models for improved and customized cyber threat intelligence
US11783233B1 (en) 2023-01-11 2023-10-10 Dimaag-Ai, Inc. Detection and visualization of novel data instances for self-healing AI/ML model-based solution deployment
US20240232713A1 (en) 2023-01-11 2024-07-11 Dimaag-Ai, Inc. Detection and visualization of novel data instances for self-healing ai/ml model-based solution deployment
US12541785B2 (en) 2023-03-03 2026-02-03 State Farm Mutual Automobile Insurance Company Chatbot to assist in vehicle shopping
US20240330151A1 (en) * 2023-04-03 2024-10-03 State Farm Mutual Automobile Insurance Company Generative Artificial Intelligence for Source Code Security Vulnerability Inspection and Remediation
US12524545B2 (en) * 2023-04-27 2026-01-13 Seagate Technology Llc Detection of ransomware attack at object store
US12158904B2 (en) * 2023-05-04 2024-12-03 Vijay Madisetti Method and system for protecting and removing private information used in large language models
US12306859B2 (en) 2023-05-04 2025-05-20 Vijay Madisetti Method and system for protecting and removing private information used in large language models
US20240394503A1 (en) 2023-05-25 2024-11-28 State Farm Mutual Automobile Insurance Company Providing information via a machine learning chatbot emulating traits of a person
US20240412145A1 (en) 2023-06-06 2024-12-12 State Farm Mutual Automobile Insurance Company System and method for training employees

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190005386A1 (en) 2017-07-01 2019-01-03 Intel Corporation Techniques for training deep neural networks
US20190260775A1 (en) 2018-02-19 2019-08-22 Cisco Technology, Inc. Refining synthetic malicious samples with unlabeled data
WO2019207770A1 (ja) 2018-04-27 2019-10-31 日本電気株式会社 学習済みモデル更新装置、学習済みモデル更新方法、プログラム
US20200125943A1 (en) 2018-10-18 2020-04-23 International Business Machines Corporation Adversial deep neural network fuzzing
WO2020165935A1 (ja) 2019-02-12 2020-08-20 日本電気株式会社 モデル構築装置、モデル構築方法、コンピュータプログラム及び記録媒体
US20210012188A1 (en) 2019-07-09 2021-01-14 Baidu Usa Llc Systems and methods for defense against adversarial attacks using feature scattering-based adversarial training

Family Cites Families (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160055240A1 (en) * 2014-08-22 2016-02-25 Microsoft Corporation Orphaned utterance detection system and method
US11526601B2 (en) * 2017-07-12 2022-12-13 The Regents Of The University Of California Detection and prevention of adversarial deep learning
US11036857B2 (en) * 2018-11-15 2021-06-15 International Business Machines Corporation Protecting a machine learning model
US11227215B2 (en) * 2019-03-08 2022-01-18 International Business Machines Corporation Quantifying vulnerabilities of deep learning computing systems to adversarial perturbations
WO2020205597A1 (en) * 2019-03-29 2020-10-08 Intel Corporation Autonomous vehicle system
JP7175244B2 (ja) * 2019-07-30 2022-11-18 Kddi株式会社 分類装置、学習装置、分類方法及びプログラム
US11856276B2 (en) * 2019-09-13 2023-12-26 Oracle International Corporation Scalable architecture for automatic generation of content distribution images
US11256754B2 (en) * 2019-12-09 2022-02-22 Salesforce.Com, Inc. Systems and methods for generating natural language processing training samples with inflectional perturbations
US11765202B2 (en) * 2020-01-17 2023-09-19 Cisco Technology, Inc. Automatic detection and prevention of injection attacks
US11544472B2 (en) * 2020-01-17 2023-01-03 Samsung Electronics Co., Ltd. Structured adversarial, training for natural language machine learning tasks
US11675896B2 (en) * 2020-04-09 2023-06-13 International Business Machines Corporation Using multimodal model consistency to detect adversarial attacks
US12088621B2 (en) * 2020-09-18 2024-09-10 Paypal, Inc. Meta-learning and auto-labeling for machine learning
CA3133729A1 (en) * 2020-10-08 2022-04-08 Royal Bank Of Canada System and method for machine learning fairness test
EP4334832A1 (en) 2021-05-07 2024-03-13 Oracle International Corporation Variant inconsistency attack (via) as a simple and effective adversarial attack method
US12524630B2 (en) * 2021-08-31 2026-01-13 Naver Corporation Adversarial generation method for training a neural model

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190005386A1 (en) 2017-07-01 2019-01-03 Intel Corporation Techniques for training deep neural networks
US20190260775A1 (en) 2018-02-19 2019-08-22 Cisco Technology, Inc. Refining synthetic malicious samples with unlabeled data
WO2019207770A1 (ja) 2018-04-27 2019-10-31 日本電気株式会社 学習済みモデル更新装置、学習済みモデル更新方法、プログラム
US20200125943A1 (en) 2018-10-18 2020-04-23 International Business Machines Corporation Adversial deep neural network fuzzing
WO2020165935A1 (ja) 2019-02-12 2020-08-20 日本電気株式会社 モデル構築装置、モデル構築方法、コンピュータプログラム及び記録媒体
US20210012188A1 (en) 2019-07-09 2021-01-14 Baidu Usa Llc Systems and methods for defense against adversarial attacks using feature scattering-based adversarial training

Also Published As

Publication number Publication date
US20220358225A1 (en) 2022-11-10
WO2022235353A1 (en) 2022-11-10
CN117296058A (zh) 2023-12-26
JP2026035651A (ja) 2026-03-04
JP2024518416A (ja) 2024-05-01
EP4334832A1 (en) 2024-03-13
US20240403447A1 (en) 2024-12-05
US12086274B2 (en) 2024-09-10
US12705368B2 (en) 2026-08-11

Similar Documents

Publication Publication Date Title
US12299402B2 (en) Techniques for out-of-domain (OOD) detection
JP7561836B2 (ja) 自然言語処理のためのストップワードデータ拡張
JP7703667B2 (ja) 固有表現認識モデルを用いたコンテキストタグ統合
JP7721559B2 (ja) 自然言語処理のためのノイズデータ拡張
US12288550B2 (en) Framework for focused training of language models and techniques for end-to-end hypertuning of the framework
US12086274B2 (en) Variant inconsistency attack (VIA) as a simple and effective adversarial attack method
JP7789778B2 (ja) 自然言語処理のためのドメイン外データ拡張
US12512091B2 (en) Fine-tuning multi-head network from a single transformer layer of pre-trained language model
JP7828346B2 (ja) 自然言語処理のためのキーワードデータ拡張ツール
JP2025166050A (ja) 自然言語処理のための強化されたロジット
KR102821062B1 (ko) 사전-트레이닝된 언어 모델들에 대한 긴 텍스트를 핸들링하기 위한 시스템 및 기술들
CN114556322A (zh) 用于定义机器学习(ml)解决方案的聊天机器人
JP2025528391A (ja) 名前付きエンティティ認識モデルの訓練を容易にするための適応的訓練データ拡大

Legal Events

Date Code Title Description
A521 Request for written amendment filed

Free format text: JAPANESE INTERMEDIATE CODE: A523

Effective date: 20241216

A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20241216

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20251021

TRDD Decision of grant or rejection written
A01 Written decision to grant a patent or to grant a registration (utility model)

Free format text: JAPANESE INTERMEDIATE CODE: A01

Effective date: 20251028

A61 First payment of annual fees (during grant procedure)

Free format text: JAPANESE INTERMEDIATE CODE: A61

Effective date: 20251118

R150 Certificate of patent or registration of utility model

Ref document number: 7778160

Country of ref document: JP

Free format text: JAPANESE INTERMEDIATE CODE: R150