Conference Presentations

Automatic IFT for Live Risk Transition

QuantMinds presentation demonstrating how Automated Implicit Function Theorem enables transition from overnight batch risk to real-time Live Risk without multi-year IT transformation.

Key Takeaways

Transform overnight batch risk into real-time Live Risk
Automatic differentiation of calibration routines using AIFT
No hidden variable tracking or large Jacobian computation required
QuantLib integration achieving 20ms pricing + Greeks for 1000 swaps

Technical Highlights

Automated Implicit Function Theorem (AIFT) for solver differentiation
Works with exact-fit and nearly-exact calibrations
No codebase refactoring required
Checkpointing approach without going into solvers
Published in Risk.NET 2022 by Goloubentsev, Lakshtanov, and Piterbarg

WBS Talk: AAD for Longstaff-Schwartz American Monte Carlo

Conference talk on AAD for Longstaff-Schwartz American Monte Carlo with memory-efficient implementation.

Key Takeaways

Pathwise + regression AAD
Memory only 2x primal
Better stability than bumping

WBS2026: Model validation from the AADC tape

Dmitri Goloubentsev

Presented at the 22nd WBS Quantitative Finance Conference in Valletta. From one recording of an unmodified QuantLib swaption: 140 constant sites, 85 frozen branches, dependencies, and where a replay stops being the model. Every number reproduces from pip install aadc-quantlib-tracing.

Key Takeaways

What an AADC tape records, attributed to source file and line
Finding hardcoded constants and frozen branches in an unmodified QuantLib pricer
Replay that flags where a recording stops being valid
Validation evidence laid out against PRA SS1/23, SR 26-2 and the ECB guide

Technical Highlights

74,428 tape entries, 100% attributed, recorded in 17 ms
Three swaption engines and five barrier engines, each on its own tape
AAD sensitivities against central finite differences to 3e-10 of total risk

WBS2025: Greeks for Autocallables

Dmitri Goloubentsev

Presentation from the 21st WBS Quantitative Finance Conference in Palermo demonstrating production-ready autocallable Greeks calculation with 99.8% cloud cost reduction.

Key Takeaways

Production-ready autocallable Greeks methodology
How to achieve 99.8% cloud cost reduction
Smoothing techniques for discontinuous payoffs
Complete open-source implementation

Technical Highlights

Pathwise Greeks with smoothing
Automatic Adjoint Differentiation
Open-source methodology

ING Analysis: Cross-Gamma Performance Comparison

Stephan Bosch

ING comparison of bumping-over-AAD for second-order cross-gamma (CVA portfolio).

Key Takeaways

Scalability of tape vs code-gen AAD for large Hessians

Technical Highlights

Code-gen linear vs tape quadratic
Bumping-over-AAD unlocks massive reuse

ING Presentation: AAD Engine Scalability Analysis

Stephan Bosch, ING

ING presentation comparing AAD engine scalability for Monte Carlo and higher-order sensitivities.

Key Takeaways

Linear vs quadratic scaling in AAD engines
When code-generation wins

Technical Highlights

Code-generation AAD scales linearly
Vectorisation critical for MC

WBS2023: Expected Shortfall with AAD

Svetlana Borovkova

Presentation on Expected Shortfall and AAD computational costs.

Key Takeaways

AAD application to regulatory risk metrics

Conference Presentations

Talks & Workshops

Automatic IFT for Live Risk Transition

Key Takeaways

Technical Highlights

WBS Talk: AAD for Longstaff-Schwartz American Monte Carlo

Key Takeaways

WBS2026: Model validation from the AADC tape

Key Takeaways

Technical Highlights

WBS2025: Greeks for Autocallables

Key Takeaways

Technical Highlights

ING Analysis: Cross-Gamma Performance Comparison

Key Takeaways

Technical Highlights

ING Presentation: AAD Engine Scalability Analysis

Key Takeaways

Technical Highlights

WBS2023: Expected Shortfall with AAD

Key Takeaways

Learn More

How AADC Works

Performance Benchmarks

Interested in Speaking Engagements?