8:50 Chair’s opening remarks

Chris SmigielskiModel Risk Director, Arvest Bank


9:00 Reviewing the expanding scope and definition of model risk and increased validation requirements

Session details 

  • Scope of model risk and validation work expanding
  • Depth of analysis required from regulators
  • Trend towards outsourcing aspects of model risk
    • Inclusion of validation in outsourcing work
  • Maintaining resources with increasing demand
  • Drivers behind resource constraints and scaling up internally
  • Inclusion of qualitative models under model risk

Olga CollinsED, Global Head of Model Risk Infrastructure and Reporting, Morgan Stanley
Irfan KaziDirector, Advanced Analytics, Internal Audit, CIBC.


9:50 Maximizing your AI/ML RoI with a balanced resourcing and deployment strategy

Session details 

  • Leveraging AI/ML for effective model development, strategy design and optimization
  • Demonstrating a compelling return on investment
  • Balancing insourcing and outsourcing analytics resources, programs and services
  • Turning advanced analytics projects and programs into a business utility
    • Realizing increased operational efficiencies, speed to market and growth strategy

Kathleen Maley, Vice President, Analytics Product Management, Experian

10:30 Morning refreshment break and networking


11:00 Managing proliferation of AI and machine learning models and understanding changing risk dynamic

Session details 

  • Speed of development and deployment
  • Application beyond traditional model scope
  • Impact on control standards with increased use
  • Recognition of process constraints and risks
  • Accepting lower control and development standards to stay ahead of competition
  • Validation requirements
    • Validation frequency
    • Automating validation
  • Regulatory framework for treatment of AI/ML models
    • Application of old concepts to new models
  • Model validation and model lifecycle: Developing standardised methodology
  • Dynamic adaptation to new data
  • Managing reputational risk of model output
  • Scaling up model production in a fast paced environment

David PalmerSenior Supervisory Financial Analyst, Federal Reserve Board
Irfan KaziDirector, Advanced Analytics, Internal Audit, CIBC


12:00 Managing ethical considerations and removing bias from data and model output

Session details 

  • Understanding how models are trained
  • Imitation of human decisioning
  • Feeding unconscious bias into machines
  • Managing risk of bias in models
  • Managing bias to protected characteristics
  • Developing tests and structures into models to demonstrate to regulators
  • Ethics of decision making
  • Transparency around decision making and ethics
  • Producing socially acceptable outcomes

Imir Arifi, Head of Methodologies and Models in the Americas, UBS 

12:40 Lunch break and networking


1:40 Model Risk Management Process Automation – Practitioner Insights

Session details 

  • Direct and indirect ways that the pandemic has impacted Model Risk Management (e.g. model malfunctioning, increased monitoring, talent scarcity)
  • Model Risk Management processes which most benefit from automation
  • Automation is not a one-size-fits-all: different automation strategies, where they succeed or fail

Steve Lindo, Course Designer and Lecturer, Columbia University MS in Enterprise Risk Management
Manoj Singh, MD and Model Risk Officer, Bank of America
Rick Boesch, Head of Model Risk Management Automation, Evalueserve


2:20 Reviewing the evolution of model risk and scoping future decisions

Session details 

  • Fintech as a driver within model risk
  • Behavioural models and uses within areas including financial crime
  • Direction of model risk towards machine learning
  • Definitions and scope of global regulation
  • Evolution of economies and political environments and impact on models
    • Foundational capabilities to evolve quickly
  • Impact of legacy infrastructures on evolution
  • Impact of remote working and resourcing skill gap
  • Maintaining proprietary models in remote environment
  • Impact of Covid on modeling
    • Treatment of extreme events
    • Developing models for future resilience

Chris SmigielskiModel Risk Director, Arvest Bank
Saqib JamshedMD, Model Risk Management, The OCC
Liming Brotcke, Senior Director, Head of Model Validation, Ally
Dave Trier, Vice President of Product, ModelOp
Stephen Hsu, SVP, Head of Model Risk Management, Pacific Western Bank 

3:10 Afternoon refreshment break and networking

3:40 Models and algorithms in Broker-Dealer – Challenges and solutions

Session details 

  • New and traditional applications within BD
  • Regulatory requirements in the industry across Europe and the US
  • Expectations for model risk management
  • Successful strategies for model automation
  • Outlook and challenges ahead

Julian HorkyHead of Risk Controlling, Berenberg Capital Markets


4:20 Quantifying climate risk

Session details 

  • Literature review of 2021 FRB NY Climate Stress Testing paper
  • Case studies of practical approaches in incorporating climate risk in:
    • Credit Underwriting
    • Portfolio Stress Testing and ACL CECL

Arsa OemarDirector, MUFG Union Bank


5:00 Modeling climate risk and determining exposure to ESG

Session details 

  • Identifying models to use
  • Determining approach with internal or external models
  • Feedback and expectation from regulators
  • Building independent models
  • Defining climate risk to advance model processes

Meet Shah, Executive Director, Operational Risk, UBS
Moez Hababou, Director, Model Risk Management, BNP Paribas
Nav Vaidhyanathan, Head of Model Risk Management, M&T Bank

5:40 Chair’s closing remarks

5:50 End of day one and drinks reception

8:50 Chair’s opening remarks

Mike Guglielmo, Managing Director, Darling Consulting Group


9:00 Establishing a strong risk culture to manage AI and machine learning models

Session details 

  • Culture to manage development and deploying of models
  • Developing culture to attract talent
  • Implementing a technology culture
  • Evolution of models and treatment within financial services
  • Incorporating model risk at the heart of risk management

Liming Brotcke, Senior Director, Head of Model Validation, Ally


9:35 Incorporating explainability and interpretability into AI and machine learning models

Session details 

  • Regulatory expectation of explainability
  • Conceptual soundness evaluation
  • Explainable machine learning
  • Designing inherently interpretable models
  • Incorporating constraints
  • Adverse action reason code

Agus Sudjianto, EVP, Head of Corporate Model Risk, Wells Fargo 

10:10 Navigating the 'black box': leading validation strategies for evolving BSA/AML models

Session details 

  • Introduction to the importance of BSA/AML models
  • Types of BSA/AML models: rules-based vs. decision-making
  • Use of machine learning and AI techniques in BSA/AML models
  • Evolution of model risk management as modeling techniques get more complicated
  • Ensuring high data quality and eliminating biases
  • New validation techniques for machine learning and AI models applied to BSA/AML
  • Case study: navigating a challenging “black box” suspicious activity monitoring model validation

Sam Chen, Quantitative Consultant, Darling Consulting Group

10:45 Morning refreshment break and networking


11:15 Integration of AI and machine learning tools under model risk management for effective validation

Session details 

  • Recalibrating AI models
  • Frequency of validation
  • Managing model risk to accommodate changes to validation
  • Understanding models in order to validate them
    • Increased use across risk siloes
  • Parameter and algorithm tuning
  • Testing bias in data sets and model inputs

Seyhun Hepdogan, Director, Model Risk Management, Discover Financial Services 
Michael Jacobs, Lead Modeling & Quantitative Analytics Expert, PNC Financial Services
Shawn Tumanov, Director, BMO Financial Group
Anantha Sharma, Director – Innovation & Data Science, Synechron 


12:00 Reviewing the evolution of expectations on staff and retention of talent in a competitive market

Session details 

  • Changes to work force with remote and hybrid environments
  • Limitations of institution dependent policies
  • Attracting academic talent with machine learning expertise
  • Mitigating talent migration and staying competitive
  • Quantifying value added by workforce
  • Talent to understand and use tools available in the market

Wei Ma, Head of Model Risk and Validation, Sumitomo Mitsui Banking Corp

12:35 Lunch break and networking


1:35 Validation considerations for AI models

Session details 

  • AI model overview
  • Predictive uncertainty
  • Model robustness
  • Perspectives on interpretability and explainability
  • Perspectives on fairness

Greg Kirczenow, Senior Director, AI Model Risk Management, RBC


2:10 Borrower level models for stress testing corporate probability of default and the quantification of model risk

Session details 

  • Introduction and motivation: The importance of quantifying model risk for stress testing
  • Review of relevant literature and industry practice
  • Hazard rate model framework for stress testing of wholesale obligor PDs
  • The principle of relative entropy for measurement of model risk
  • Modeling data and econometric results
  • Model risk bounds on CECL forecast scenarios

Michael Jacobs, Lead Modeling & Quantitative Analytics Expert, PNC Financial Services

2:45 Afternoon refreshment break and networking

3:15 Model risk and model risk audit – Auditing model risk management

Session details 

  • Three lines of defense model and Internal Audit’s role specified in SR 11-7
  • Testing and Evaluating Model Risk Governance
  • Testing and Evaluating Model Development Process
  • Testing and Evaluating Model Validation Process
  • Validating Remediation of Regulatory Matters
  • Aggregation of Audit Findings and SR 11-7 Gap Analysis
  • Challenges in Model Risk Management and Model Risk Audit

Jane Zheng, Director, Model Risk Audit, BNY Mellon


3:50 Automating model risk management activities to increase efficiency

Session details 

  • Automating validation on an ongoing basis
  • Model governance to monitor models
  • Ongoing performance monitoring of models
  • Enhancing efficiency of model risk teams
    • Reducing human intervention on manual tasks
  • Leveraging AI to automate tasks whilst retaining human oversight
  • Identifying areas that can be automated effectively

Jing Zou, Managing Director, Model Risk Management, Royal Bank of Canada
Mitchell Button, SVP, Model Validation, US Bank 
Florentino Rico, Senior Manager, Model Risk Management, Discover Financial Services
Wei Ma, Head of Model Risk and Validation, Sumitomo Mitsui Banking Corp
Andreza Barbosa,
Global Head of Model Governance, Goldman Sachs 

4:35 Chair’s closing remarks and end of Congress