SlideShare ist ein Scribd-Unternehmen logo
1 von 15
IBM Research – Tokyo




Opportunistic Adversaries
– On Imminent Threats to
Learning-based Business Automation –

                       Michiaki Tatsubori, IBM Research – Tokyo
                       Shohei Hido, Preferred Infrastructure, Inc.




M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012   © 2012 IBM Corporation
IBM Research – Tokyo




About This Talk

    § A business process with automated decision through
      machine learning is useful & promising


    § The “opportunistic adversaries” – potential adversaries
      exploiting its misclassification, which is inevitable
       – A case study with loan exam automation


    § A reference design & implementation of counter
      measures


2       M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012   © 2012 IBM Corporation
IBM Research – Tokyo



Business Processes with Machine Learning
– a Promising Approach
                                                                An Example of Credit Card Fraud Detection
                          BPM                                                      ML
                 Order
               Validation

                                         Order parameter                   Transparent
                                                                            Transparent
                 Fraud                                                        Decision
                                                                               Decision
               Detection                Report parameter                       Service
                                                                                Service
                                          (e.g. exception)
                                                                                                            Models
                                                                                                             Models
    Exception?                  Exception
                       Yes      handling
                 No                                                                                  Induce models



     No                         Yes
                 Order
               accepted?                                                      Learning
                                        Training parameter                     Learning
                                                                               Service
                                                                                Service
                                         & decision record                                                    History
                                                                                                               History
                           Process                                                                           Repository
                                                                                                              Repository
      Review
                             Order
                                                                                                  Order process histories
                                                                                                 Order rejection histories
3           M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose          Jul 25, 2012          © 2012 IBM Corporation
IBM Research – Tokyo




Potential Application: Loan Exam Processing




4      M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012   © 2012 IBM Corporation
IBM Research – Tokyo




Potential Application: Insurance Claims Processing




5      M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012   © 2012 IBM Corporation
IBM Research – Tokyo




Supervised Machine Learning is the Key Technology
                                                      D training = {( x1 , y1 ),..., (x n , yn )}

§ Machine learning for                                where x i ∈ V (V : feature - vector space)
                                                      and y j ∈ C (C : a set of class labels)
  process automation:
    – Learning from known                                Learning Data:
      decisions for input
                                                                                Approve
      parameters
                                                                                       Distinction by
    – Allowing automated                                                          a ground-truth function
                                                                                        (unknown)
      decision for unknown input
      parameters                                                                                                     Models
                                                                                                                      Models

    Ex. Insurance claim
     processing, credit order
     approval, etc.                                               Reject



6        M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose              Jul 25, 2012     © 2012 IBM Corporation
IBM Research – Tokyo


Inevitable Misclassifications are Compensated
by Other Benefits
                                                                     Produce a function
    § Hard to avoid misclassifications                               h:x → y
       – Tradeoffs between false                                     where x ∈ V (V : feature - vector space)
         positives versus false                                      and y ∈ C (C : a set of class labels)
         negatives                                                       Test Data:
    § Overall business models can                                                         Approve
      compensate loss from
                                                                                        FP           Distinction by
      misclassifications with benefit                                                             a learned function
                                                                                                     (probabilistic)
      from automation:
       – Less human workload
       – Less careless misses
                                                                                   FN

                                                                               Reject

7       M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose       Jul 25, 2012               © 2012 IBM Corporation
IBM Research – Tokyo




Opportunistic Adversaries: Threats by Adversaries
Outsmarting Machinery Misjudgment
    – Opportunistic adversaries                                             Test Data:
      scenario:
                                                                                               Approve
      • A user detects the
        misclassification by the                                                         FP
                                                                                          FP FP
                                                                                           FP
                                                                                            FP
        system for certain input
        parameters
      • Attackers provide parameters
        so that they resemble the
        former input parameters                                                       FN
        misclassified
    Ex. A manual for “legally                                                    Reject
     cheating insurance claims”

8         M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose      Jul 25, 2012     © 2012 IBM Corporation
IBM Research – Tokyo


Conditions Where Opportunistic Adversaries Become
Threats

    § Threat: Damages from spreading adversaries which outsmart inevitable
      false positives/negatives with ML, under the condition:
        – Attacks intentionally forge inputs (integrity attack),
        – Attacks start from a tiny false positive/negative case revealed to
          potential attackers (exploratory and indiscriminate attack), and
        – Unawareness of damages (stealthy attack)


    § Existing works didn’t address this situations or required impractical
      amount of learning and test samples
       – Transfer learning [Sugiyama 2006]
       – Adversarial learning [Lowd 2005]
       – Outlier detection [Hido 2008]




9        M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012   © 2012 IBM Corporation
IBM Research – Tokyo

BPM & Abstract Decision Service + Anomaly Detection

                           BPM                                                      Decision Service
                  Order
                Validation

                                          Order parameter                   Transparent
                                                                             Transparent
                  Fraud                                                        Decision
                                                                                Decision
                Detection                Report parameter                       Service
                                                                                 Service
                                           (e.g. exception)                                               Models
                                                                                                           Models
     Exception?                  Exception
                        Yes      handling
                  No
                                                                                                          Rule
                                                                                                           Rule
                                                                                                        Repository
                                                                                                         Repository
      No                         Yes
                  Order
                accepted?

                                                                             History of
                                                                              History of
                            Process                                          Automated
                                                                              Automated
       Review                                                                Decisions
                              Order                                           Decisions               Input Frequencies
                                                                                                       Input Frequencies




10           M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose      Jul 25, 2012           © 2012 IBM Corporation
IBM Research – Tokyo




Reference Countermeasure Prototype Outline
     § Record timestamps of training and test                                Record timestamps of input data:
       inputs
                                                                                                A1
     § Cluster training inputs to segmentalize                           Class A
       the input space into subclasses                                                          A2
                                                                                                A3
     § Maintain frequency statistics about
       per-subclass probabilities of training                                                                                            time
       inputs for various times and timeframes                               Class B
       and test inputs for recent times and
       timeframes
     § Detect significant relative increases in
                                                                              distribution              Time series analysis
       each subclass as anomaly to alert (telling                                                       for subclass probabilities
       as an exception)
                                                                                            1   2   3    log t
        – Sensing potential attacks
           outsmarting the trained model                                    Score :
        – Giving a chance of human review                                                                            Ps( test ) (l )
           and model update                                                q( x   (test)
                                                                                           )=
                                                                                              E ( Pt ( training ) (l )) (σ ( Pt ( training) (l )) + 1)
                                                                                  k


                                                                                     where s = t k and l = g( xk )
                                                                                                 (test)        (test)




11            M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose                   Jul 25, 2012               © 2012 IBM Corporation
IBM Research – Tokyo




Architecture of Reference Implementation
              Training Input                                                         Test Input



      label
               A        B       B      A
                                                                             Time
      time
               t1       t2     t3      t4                                   stamp
                                                                                     s1     s2    s3           s4
     stamp
                                                                                                                                               Classification Output



                         Classifier                                                                  Classifier                                        A      B
                                                                                                                                                 B
                         Generator                                                                                                               s1    s2     s3
                                                             classifier


                      Sub-classifier                                                                                      Time Series
                       Generator                                                                                           Analyzer
                                                                                                                          (Test Data)
                                                      sub-classifier
                        Time Series
                         Analyzer                                                                      distribution
                                                     distribution
                                                                                                                                       frequency
                      (Training Data)
                                                                                                                 1    2   3   log t
                                                                                                                                        statistics
                                                               1    2   3    log t
                                                                                                                                       (test data)
                                               frequency                                  Anomaly
                                                statistics                                detector                    notify anomaly
                                            (training data)

12                  M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose                                              Jul 25, 2012           © 2012 IBM Corporation
IBM Research – Tokyo




Preliminary Experimental Results
§ Observed effectiveness in an                                                                              Learning Data:
                                                                                                                                 Attack
  experiment with spam filtering
     – Experimented with
       Spambase (mails with
       some spams) in UCI data                                                                              Test Data:
                                                                                                                               Clusters


     – Used first 80% for training
       and last 20% for testing
     – Replaced 5% of testing
       data with misclassified                                                                                                 Clusters



                                                                                 Freq. Ratio / Std. Dist.
       samples
     – Observed they are                                                                                                           Detected
       detected as anomaly

13        M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose                                  Jul 25, 2012              © 2012 IBM Corporation
IBM Research – Tokyo




Concluding Remarks

     § Defined “Opportunistic Adversaries”
       as a threat to automated business                                                                 Approve
       processes with machine learning                                                              FP
                                                                                                     FP FP
                                                                                                      FP
                                                                                                       FP
        – Integrity, exploratory, indiscriminate,
          and stealthy attacks

                                                                                                   FN

     § A reference solution architecture                                                       Reject
       proposed
        – + anomaly detection in temporal input
          space distribution statistics

14       M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012            © 2012 IBM Corporation
IBM Research – Tokyo




Thank you!
Questions?




M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose   Jul 25, 2012   © 2012 IBM Corporation

Weitere ähnliche Inhalte

Ähnlich wie Opportunistic Adversaries - On Imminent Threats to Learning-based Business Automation (presentation at SRII 2012)

Ian Robinson Testable Foundations
Ian  Robinson    Testable  FoundationsIan  Robinson    Testable  Foundations
Ian Robinson Testable FoundationsSOA Symposium
 
Testing Rich Domain Models
Testing Rich Domain ModelsTesting Rich Domain Models
Testing Rich Domain ModelsChris Richardson
 
Dirk Krafzig Enterprise S O Aand Dependency Mngt
Dirk  Krafzig    Enterprise S O Aand Dependency MngtDirk  Krafzig    Enterprise S O Aand Dependency Mngt
Dirk Krafzig Enterprise S O Aand Dependency MngtSOA Symposium
 
NG BB 19 Document and Analyze the Process
NG BB 19 Document and Analyze the ProcessNG BB 19 Document and Analyze the Process
NG BB 19 Document and Analyze the ProcessLeanleaders.org
 
Hyp01 essbase+planning
Hyp01 essbase+planningHyp01 essbase+planning
Hyp01 essbase+planningAmit Sharma
 
JIRA Studio: Development in the Cloud - Atlassian Summit 2010
JIRA Studio: Development in the Cloud - Atlassian Summit 2010JIRA Studio: Development in the Cloud - Atlassian Summit 2010
JIRA Studio: Development in the Cloud - Atlassian Summit 2010Atlassian
 
Discovering Concurrency: Learning (Business) Process Models from Examples
Discovering Concurrency: Learning (Business) Process Models from ExamplesDiscovering Concurrency: Learning (Business) Process Models from Examples
Discovering Concurrency: Learning (Business) Process Models from ExamplesWil van der Aalst
 
Process Mining - Chapter 1 - Introduction
Process Mining - Chapter 1 - IntroductionProcess Mining - Chapter 1 - Introduction
Process Mining - Chapter 1 - IntroductionWil van der Aalst
 
Process mining chapter_01_introduction
Process mining chapter_01_introductionProcess mining chapter_01_introduction
Process mining chapter_01_introductionMuhammad Ajmal
 
IdealECP presentation for Novo Nordisk
IdealECP presentation for Novo NordiskIdealECP presentation for Novo Nordisk
IdealECP presentation for Novo Nordiskcbiddle2
 
Managed services. preparing for market convergence (mswc 2011)
Managed services. preparing for market convergence (mswc 2011)Managed services. preparing for market convergence (mswc 2011)
Managed services. preparing for market convergence (mswc 2011)raulzamorano
 
Ventana Systems Uk
Ventana Systems UkVentana Systems Uk
Ventana Systems UkAndy Hill
 
TH e-GIF on SOA Using Open Enterprise Architecture
TH e-GIF on SOA Using Open Enterprise ArchitectureTH e-GIF on SOA Using Open Enterprise Architecture
TH e-GIF on SOA Using Open Enterprise ArchitectureThanachart Numnonda
 
Performance Testing
Performance TestingPerformance Testing
Performance TestingCodelattice
 
02 spc訓練教材
02 spc訓練教材02 spc訓練教材
02 spc訓練教材營松 林
 
Semantic Complex Event Processing at Sem Tech 2010
Semantic Complex Event Processing at Sem Tech 2010Semantic Complex Event Processing at Sem Tech 2010
Semantic Complex Event Processing at Sem Tech 2010Adrian Paschke
 
DHW Fundamentals
DHW FundamentalsDHW Fundamentals
DHW Fundamentalskapildevang
 
2 d3.javne nabavke_neum160410
2 d3.javne nabavke_neum1604102 d3.javne nabavke_neum160410
2 d3.javne nabavke_neum160410Oracle BH
 
Third party independent test audit.
Third party independent test audit.Third party independent test audit.
Third party independent test audit.Mindtree Ltd.
 

Ähnlich wie Opportunistic Adversaries - On Imminent Threats to Learning-based Business Automation (presentation at SRII 2012) (20)

Ian Robinson Testable Foundations
Ian  Robinson    Testable  FoundationsIan  Robinson    Testable  Foundations
Ian Robinson Testable Foundations
 
Testing Rich Domain Models
Testing Rich Domain ModelsTesting Rich Domain Models
Testing Rich Domain Models
 
Dirk Krafzig Enterprise S O Aand Dependency Mngt
Dirk  Krafzig    Enterprise S O Aand Dependency MngtDirk  Krafzig    Enterprise S O Aand Dependency Mngt
Dirk Krafzig Enterprise S O Aand Dependency Mngt
 
NG BB 19 Document and Analyze the Process
NG BB 19 Document and Analyze the ProcessNG BB 19 Document and Analyze the Process
NG BB 19 Document and Analyze the Process
 
Hyp01 essbase+planning
Hyp01 essbase+planningHyp01 essbase+planning
Hyp01 essbase+planning
 
JIRA Studio: Development in the Cloud - Atlassian Summit 2010
JIRA Studio: Development in the Cloud - Atlassian Summit 2010JIRA Studio: Development in the Cloud - Atlassian Summit 2010
JIRA Studio: Development in the Cloud - Atlassian Summit 2010
 
Discovering Concurrency: Learning (Business) Process Models from Examples
Discovering Concurrency: Learning (Business) Process Models from ExamplesDiscovering Concurrency: Learning (Business) Process Models from Examples
Discovering Concurrency: Learning (Business) Process Models from Examples
 
Process Mining - Chapter 1 - Introduction
Process Mining - Chapter 1 - IntroductionProcess Mining - Chapter 1 - Introduction
Process Mining - Chapter 1 - Introduction
 
Process mining chapter_01_introduction
Process mining chapter_01_introductionProcess mining chapter_01_introduction
Process mining chapter_01_introduction
 
IdealECP presentation for Novo Nordisk
IdealECP presentation for Novo NordiskIdealECP presentation for Novo Nordisk
IdealECP presentation for Novo Nordisk
 
Managed services. preparing for market convergence (mswc 2011)
Managed services. preparing for market convergence (mswc 2011)Managed services. preparing for market convergence (mswc 2011)
Managed services. preparing for market convergence (mswc 2011)
 
Ventana Systems Uk
Ventana Systems UkVentana Systems Uk
Ventana Systems Uk
 
TH e-GIF on SOA Using Open Enterprise Architecture
TH e-GIF on SOA Using Open Enterprise ArchitectureTH e-GIF on SOA Using Open Enterprise Architecture
TH e-GIF on SOA Using Open Enterprise Architecture
 
Performance Testing
Performance TestingPerformance Testing
Performance Testing
 
02 spc訓練教材
02 spc訓練教材02 spc訓練教材
02 spc訓練教材
 
Semantic Complex Event Processing at Sem Tech 2010
Semantic Complex Event Processing at Sem Tech 2010Semantic Complex Event Processing at Sem Tech 2010
Semantic Complex Event Processing at Sem Tech 2010
 
DHW Fundamentals
DHW FundamentalsDHW Fundamentals
DHW Fundamentals
 
Solution Manager Roadmaps
Solution Manager RoadmapsSolution Manager Roadmaps
Solution Manager Roadmaps
 
2 d3.javne nabavke_neum160410
2 d3.javne nabavke_neum1604102 d3.javne nabavke_neum160410
2 d3.javne nabavke_neum160410
 
Third party independent test audit.
Third party independent test audit.Third party independent test audit.
Third party independent test audit.
 

Kürzlich hochgeladen

A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationMichael W. Hawkins
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGSujit Pal
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesSinan KOZAK
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024The Digital Insurer
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Miguel Araújo
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Drew Madelung
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Igalia
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 

Kürzlich hochgeladen (20)

A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
GenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day PresentationGenCyber Cyber Security Day Presentation
GenCyber Cyber Security Day Presentation
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Google AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAGGoogle AI Hackathon: LLM based Evaluator for RAG
Google AI Hackathon: LLM based Evaluator for RAG
 
Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024Finology Group – Insurtech Innovation Award 2024
Finology Group – Insurtech Innovation Award 2024
 
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
Mastering MySQL Database Architecture: Deep Dive into MySQL Shell and MySQL R...
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
Strategies for Unlocking Knowledge Management in Microsoft 365 in the Copilot...
 
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
Raspberry Pi 5: Challenges and Solutions in Bringing up an OpenGL/Vulkan Driv...
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 

Opportunistic Adversaries - On Imminent Threats to Learning-based Business Automation (presentation at SRII 2012)

  • 1. IBM Research – Tokyo Opportunistic Adversaries – On Imminent Threats to Learning-based Business Automation – Michiaki Tatsubori, IBM Research – Tokyo Shohei Hido, Preferred Infrastructure, Inc. M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 2. IBM Research – Tokyo About This Talk § A business process with automated decision through machine learning is useful & promising § The “opportunistic adversaries” – potential adversaries exploiting its misclassification, which is inevitable – A case study with loan exam automation § A reference design & implementation of counter measures 2 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 3. IBM Research – Tokyo Business Processes with Machine Learning – a Promising Approach An Example of Credit Card Fraud Detection BPM ML Order Validation Order parameter Transparent Transparent Fraud Decision Decision Detection Report parameter Service Service (e.g. exception) Models Models Exception? Exception Yes handling No Induce models No Yes Order accepted? Learning Training parameter Learning Service Service & decision record History History Process Repository Repository Review Order Order process histories Order rejection histories 3 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 4. IBM Research – Tokyo Potential Application: Loan Exam Processing 4 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 5. IBM Research – Tokyo Potential Application: Insurance Claims Processing 5 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 6. IBM Research – Tokyo Supervised Machine Learning is the Key Technology D training = {( x1 , y1 ),..., (x n , yn )} § Machine learning for where x i ∈ V (V : feature - vector space) and y j ∈ C (C : a set of class labels) process automation: – Learning from known Learning Data: decisions for input Approve parameters Distinction by – Allowing automated a ground-truth function (unknown) decision for unknown input parameters Models Models Ex. Insurance claim processing, credit order approval, etc. Reject 6 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 7. IBM Research – Tokyo Inevitable Misclassifications are Compensated by Other Benefits Produce a function § Hard to avoid misclassifications h:x → y – Tradeoffs between false where x ∈ V (V : feature - vector space) positives versus false and y ∈ C (C : a set of class labels) negatives Test Data: § Overall business models can Approve compensate loss from FP Distinction by misclassifications with benefit a learned function (probabilistic) from automation: – Less human workload – Less careless misses FN Reject 7 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 8. IBM Research – Tokyo Opportunistic Adversaries: Threats by Adversaries Outsmarting Machinery Misjudgment – Opportunistic adversaries Test Data: scenario: Approve • A user detects the misclassification by the FP FP FP FP FP system for certain input parameters • Attackers provide parameters so that they resemble the former input parameters FN misclassified Ex. A manual for “legally Reject cheating insurance claims” 8 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 9. IBM Research – Tokyo Conditions Where Opportunistic Adversaries Become Threats § Threat: Damages from spreading adversaries which outsmart inevitable false positives/negatives with ML, under the condition: – Attacks intentionally forge inputs (integrity attack), – Attacks start from a tiny false positive/negative case revealed to potential attackers (exploratory and indiscriminate attack), and – Unawareness of damages (stealthy attack) § Existing works didn’t address this situations or required impractical amount of learning and test samples – Transfer learning [Sugiyama 2006] – Adversarial learning [Lowd 2005] – Outlier detection [Hido 2008] 9 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 10. IBM Research – Tokyo BPM & Abstract Decision Service + Anomaly Detection BPM Decision Service Order Validation Order parameter Transparent Transparent Fraud Decision Decision Detection Report parameter Service Service (e.g. exception) Models Models Exception? Exception Yes handling No Rule Rule Repository Repository No Yes Order accepted? History of History of Process Automated Automated Review Decisions Order Decisions Input Frequencies Input Frequencies 10 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 11. IBM Research – Tokyo Reference Countermeasure Prototype Outline § Record timestamps of training and test Record timestamps of input data: inputs A1 § Cluster training inputs to segmentalize Class A the input space into subclasses A2 A3 § Maintain frequency statistics about per-subclass probabilities of training time inputs for various times and timeframes Class B and test inputs for recent times and timeframes § Detect significant relative increases in distribution Time series analysis each subclass as anomaly to alert (telling for subclass probabilities as an exception) 1 2 3 log t – Sensing potential attacks outsmarting the trained model Score : – Giving a chance of human review Ps( test ) (l ) and model update q( x (test) )= E ( Pt ( training ) (l )) (σ ( Pt ( training) (l )) + 1) k where s = t k and l = g( xk ) (test) (test) 11 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 12. IBM Research – Tokyo Architecture of Reference Implementation Training Input Test Input label A B B A Time time t1 t2 t3 t4 stamp s1 s2 s3 s4 stamp Classification Output Classifier Classifier A B B Generator s1 s2 s3 classifier Sub-classifier Time Series Generator Analyzer (Test Data) sub-classifier Time Series Analyzer distribution distribution frequency (Training Data) 1 2 3 log t statistics 1 2 3 log t (test data) frequency Anomaly statistics detector notify anomaly (training data) 12 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 13. IBM Research – Tokyo Preliminary Experimental Results § Observed effectiveness in an Learning Data: Attack experiment with spam filtering – Experimented with Spambase (mails with some spams) in UCI data Test Data: Clusters – Used first 80% for training and last 20% for testing – Replaced 5% of testing data with misclassified Clusters Freq. Ratio / Std. Dist. samples – Observed they are Detected detected as anomaly 13 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 14. IBM Research – Tokyo Concluding Remarks § Defined “Opportunistic Adversaries” as a threat to automated business Approve processes with machine learning FP FP FP FP FP – Integrity, exploratory, indiscriminate, and stealthy attacks FN § A reference solution architecture Reject proposed – + anomaly detection in temporal input space distribution statistics 14 M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation
  • 15. IBM Research – Tokyo Thank you! Questions? M Tatsubori & S Hido: Opportunistic Adversaries, SRII 2012, San Jose Jul 25, 2012 © 2012 IBM Corporation

Hinweis der Redaktion

  1. On Imminent Threats to Learning-based Business Automation
  2. Automated business processes with machine learning technologies, where machines make decisions in place of human operators, are coming into practical use [3, 11, 12, 17]. They are being used even for mission-critical operations such as credit and loan decision making [16] and fraud detection [7, 2]. While misclassifications are hard to avoid in practice, automation-based business is still viable even with limited classification accuracy by machines, thanks to business models that can cover the losses from rare mistakes with the profits from the large savings in human costs and reductions in routine losses due to relatively reliable automation.