
Derivatives Pricing and Managing with AI & Quantum Computing
From Market Data to Real-Time Pricing, Deep Hedging and Autonomous Derivatives Desks
​
Course Overview
​
Artificial Intelligence is transforming derivatives pricing from a model-centric discipline into an integrated computational architecture combining market data intelligence, calibration, valuation, sensitivities, portfolio revaluation, hedging, model validation and automated decision support.
This programme provides a practical and forward-looking framework for pricing and managing derivatives using Machine Learning, Deep Learning, Generative AI, Reinforcement Learning, differentiable models and quantum algorithms.
Rather than treating AI as an isolated set of techniques, the course integrates AI directly into the derivatives pricing workflow:
MARKET DATA → CALIBRATION → PRICING → GREEKS → HEDGING → PORTFOLIO REVALUATION → ALL-IN PRICE → VALIDATION → AI AGENTS → QUANTUM COMPUTING
Participants will learn how traditional pricing models, numerical methods and modern AI architectures can work together to build faster, more scalable and more intelligent derivatives pricing systems.
​
Course Objectives
​
By the end of the programme, participants will be able to:
​
-
Understand the architecture of a modern AI-enabled derivatives pricing platform.
-
Build and validate market data pipelines for curves, volatility surfaces and pricing inputs.
-
Apply Artificial Intelligence to yield-curve and volatility-surface calibration.
-
Develop neural surrogate models capable of approximating complex pricing functions at high speed.
-
Use Automatic Differentiation and differentiable programming to calculate Greeks efficiently.
-
Apply Neural SDEs, Physics-Informed Neural Networks and Neural Operators to derivatives valuation.
-
Price complex, exotic and path-dependent derivatives using advanced AI methods.
-
Implement AI-driven pricing for interest-rate, FX and cross-currency derivatives.
-
Revalue large derivatives portfolios using AI surrogate models and scenario engines.
-
Apply Reinforcement Learning and Deep Hedging to dynamic hedging problems.
-
Understand how CVA, DVA, FVA, MVA and collateral affect the all-in price of a derivative.
-
Use AI to optimise Initial Margin, SIMM and collateral allocation.
-
Apply Generative AI to term-sheet interpretation, model selection, documentation and pricing workflows.
-
Design multi-agent architectures for derivatives pricing and risk management.
-
Validate AI pricing models and quantify model uncertainty.
-
Understand the role of Quantum Amplitude Estimation, Quantum Monte Carlo and hybrid quantum-classical optimisation in derivatives.
-
Compare classical, AI and quantum methods in terms of accuracy, speed, robustness and implementation constraints.
Who Should Attend?
This programme is designed for professionals involved in derivatives valuation, quantitative finance, trading, risk management and model governance.
It is particularly relevant for:
-
Derivatives Traders and Structurers
-
Quantitative Analysts and Quant Developers
-
Front-Office and XVA Quants
-
Market Risk Professionals
-
Model Validation and Model Risk Teams
-
ALM and Treasury Professionals
-
Counterparty Credit Risk Specialists
-
Risk Analytics and Data Science Teams
-
Financial Engineers
-
Internal Audit and Independent Validation Teams
-
Regulators and Supervisors
-
Consultants specialising in derivatives, valuation and quantitative risk
-
Professionals working on AI applications in financial markets
A working knowledge of derivatives, probability and basic quantitative finance is recommended. Experience with Python, R or numerical methods is useful but not essential for understanding the programme architecture.


AGENDA
Derivatives Pricing and Managing with AI and
Quantum Computing
Module 1 — The AI-Powered Derivatives Pricing Architecture
From Trade to Real-Time Valuation
​
This module introduces the complete derivatives pricing workflow and shows where Artificial Intelligence creates the greatest value across market data, calibration, pricing, sensitivities, portfolio revaluation and model validation.
Key Topics
-
Modern derivatives pricing architecture
-
Trade representation
-
Market data
-
Pricing libraries
-
Model selection
-
Calibration workflows
-
Real-time vs batch pricing
-
CPU and GPU computation
-
AI surrogate models
-
Model governance
-
Pricing controls
-
Production implementation
RiskLab 1 — Build an AI Derivatives Pricing Factory
Build a pricing workflow that receives a derivative trade and automatically returns:
Price + Greeks + Pricing Model + Confidence Measure + Computation Time
Module 2 — AI Market Data Intelligence & Arbitrage-Free Data
​
Building Reliable Inputs for Pricing Engines
Derivatives pricing is only as reliable as the market data feeding the models. This module applies Artificial Intelligence to detect anomalies, stale quotes, missing information and arbitrage inconsistencies.
Key Topics
-
Market data quality
-
Tick and quote data
-
Bid-ask spreads
-
Missing observations
-
Stale prices
-
Outlier detection
-
Isolation Forest
-
Autoencoders
-
Time-series anomaly detection
-
Data reconciliation
-
No-arbitrage constraints
-
Market data lineage
RiskLab 2 — AI Market Data Cleaner
Detect and repair anomalies across:
Interest Rates + Swaps + Options + Volatility Quotes
before feeding the data into the pricing engine.
Module 3 — AI Multi-Curve & Interest Rate Calibration
​
From SOFR Curves to Intelligent Calibration
This module develops modern curve-construction techniques combining traditional bootstrapping with AI-based calibration.
Key Topics
-
OIS discounting
-
SOFR
-
Forward curves
-
Basis curves
-
Cross-currency curves
-
Bootstrapping
-
Interpolation
-
Nelson-Siegel-Svensson
-
Newton-Raphson
-
Gaussian Processes
-
Neural calibration
-
Differentiable curve construction
RiskLab 3 — Neural Multi-Curve Calibration Engine
Compare:
Classical Bootstrapping
vs
Neural Network Calibration
using:
Accuracy | Stability | Recalibration Speed | Pricing Error
​
Module 4 — AI Volatility Surface & Arbitrage-Free Smile Modelling
​
From Market Quotes to Tradable Volatility Surfaces
The module develops volatility surfaces that are smooth, stable and consistent with no-arbitrage conditions.
Key Topics
-
Implied volatility
-
Smile and skew
-
Local volatility
-
Heston
-
SABR
-
SVI
-
Surface interpolation
-
Butterfly arbitrage
-
Calendar arbitrage
-
Neural volatility surfaces
-
Gaussian Processes
-
Generative modelling
-
Arbitrage penalties
RiskLab 4 — Arbitrage-Free Neural Volatility Surface
Train a neural model to reconstruct an incomplete volatility surface while automatically penalising arbitrage violations.
​
Module 5 — Neural Surrogate Pricing Engines
​
From Complex Numerical Models to Millisecond Pricing
AI surrogate models learn complex pricing functions and allow extremely fast subsequent revaluation.
Key Topics
-
Surrogate models
-
Universal approximation
-
Training data generation
-
Monte Carlo labels
-
Deep Neural Networks
-
Gradient Boosting
-
Approximation error
-
Portfolio revaluation
-
GPU acceleration
-
Tail accuracy
-
Extrapolation risk
RiskLab 5 — Monte Carlo vs Neural Pricing Engine
Train a neural pricing engine using Monte Carlo-generated prices and compare:
Pricing Error | Runtime | Speed-Up | Tail Error
Module 6 — Differentiable Pricing, Automatic
Differentiation & AI
​
Greeks
Price and Sensitivities in One Computational Graph
This module combines Adjoint Algorithmic Differentiation with modern differentiable programming.
Key Topics
-
Automatic Differentiation
-
Adjoint differentiation
-
Computational graphs
-
Pathwise derivatives
-
Delta
-
Gamma
-
Vega
-
Rho
-
Cross-Greeks
-
Neural Greeks
-
Portfolio sensitivities
-
Differentiable calibration
RiskLab 6 — Real-Time Neural Greeks Engine
Compare:
Finite Differences vs AAD vs Automatic Differentiation Neural Networks for large derivatives portfolios.
​
Module 7 — Neural SDEs & AI Dynamics for Derivatives
​
Learning Financial Dynamics from Data
Instead of assuming fixed parametric dynamics, AI can learn parts of the underlying stochastic process directly from market information.
Key Topics
-
Stochastic Differential Equations
-
Neural SDEs
-
Neural ODEs
-
Latent state models
-
Learned drift
-
Learned diffusion
-
Risk-neutral dynamics
-
Hybrid structural-AI models
-
Calibration
-
Simulation
-
Model constraints
RiskLab 7 — Neural SDE Option Pricing
Compare:
Heston vs Neural SDE
for fitting market dynamics and pricing options.
Module 8 — Neural PDEs & Deep Operator Learning
​
Solving Pricing Equations with Artificial Intelligence
AI techniques can approximate solutions to high-dimensional pricing equations where traditional methods become computationally expensive.
Key Topics
-
Black-Scholes PDE
-
Feynman-Kac
-
Physics-Informed Neural Networks
-
Deep BSDE
-
Neural PDE solvers
-
DeepONet
-
Fourier Neural Operators
-
Boundary conditions
-
High-dimensional pricing
-
Error analysis
RiskLab 8 — PDE vs PINN vs Neural Operator
Solve the same derivatives pricing problem using:
Finite Differences vs PINN vs Neural Operator
and compare accuracy and computational scalability.
Module 9 — AI Pricing of Exotic & Path-Dependent Derivatives
From Barriers to Autocallables
This module applies AI to derivatives where path dependency, early exercise and high dimensionality make traditional pricing expensive.
Key Topics
-
Barrier options
-
Asian options
-
Lookback options
-
Cliquets
-
Autocallables
-
Bermudan options
-
Basket options
-
Path dependence
-
Monte Carlo
-
Longstaff-Schwartz
-
Neural regression
-
Exercise strategies
RiskLab 9 — AI Autocallable & Bermudan Pricing
Compare:
-
Monte Carlo
-
Longstaff-Schwartz
-
Neural Network
for pricing and exercise strategy estimation.
​
Module 10 — AI Interest Rate Derivatives Pricing
​
Swaptions, Caps, Floors and Callable Structures
AI is applied to calibration and revaluation of complex interest-rate products.
Key Topics
-
Caps and Floors
-
Swaptions
-
Bermudan Swaptions
-
Callable bonds
-
Hull-White
-
LIBOR Market Model
-
SABR
-
SOFR
-
Neural calibration
-
Surrogate pricing
-
Dynamic Greeks
RiskLab 10 — AI Bermudan Swaption Pricing
Compare:
Tree vs Longstaff-Schwartz vs Deep Neural Network
for price and optimal exercise estimation.
Module 11 — AI FX & Cross-Currency Derivatives
​
Pricing Across Currencies, Curves and Volatility
The module integrates FX, interest-rate and basis risks within a single pricing framework.
Key Topics
-
FX forwards
-
FX options
-
Cross-Currency Swaps
-
Cross-currency basis
-
Quanto derivatives
-
Correlation
-
FX volatility surfaces
-
Multi-currency discounting
-
AI calibration
-
Scenario revaluation
RiskLab 11 — Multi-Currency AI Pricing Engine
Price and analyse:
FX Option + Cross-Currency Swap + Quanto Option
under multiple rate and volatility scenarios.
Module 12 — Portfolio-Level AI Pricing & Real-Time Revaluation
​
From One Trade to Large Derivatives Books
This module focuses on large-scale portfolio revaluation and computational efficiency.
Key Topics
-
Portfolio valuation
-
Scenario cubes
-
Incremental pricing
-
Full revaluation
-
Greeks approximation
-
Surrogate models
-
GPU batching
-
Portfolio Greeks
-
P&L Explain
-
What-if analysis
-
Real-time risk
RiskLab 12 — 100,000-Trade Revaluation Simulator
Compare:
Full Revaluation
vs
Greeks Approximation
vs
AI Surrogate Revaluation
for a large simulated derivatives portfolio.
Module 13 — Deep Hedging & Reinforcement Learning
​
From Greeks to Optimal Hedging Policies
This module applies Reinforcement Learning to hedging under realistic market frictions.
Key Topics
-
Classical Delta Hedging
-
Incomplete markets
-
Transaction costs
-
Discrete rebalancing
-
Reinforcement Learning
-
Deep Hedging
-
Policy networks
-
Risk-sensitive objectives
-
CVaR
-
Utility optimisation
-
Liquidity constraints
RiskLab 13 — Deep Hedging Trading Agent
Compare:
Black-Scholes Delta Hedge
vs
Reinforcement Learning Hedge
under:
Transaction Costs + Discrete Rebalancing + Volatility Shock
Module 14 — XVA-Aware Pricing for the Derivatives Desk
​
From Clean Price to All-In Price
The module provides the pricing-level understanding of valuation adjustments required by a derivatives desk without turning the programme into a full counterparty-risk course.
Key Topics
-
Clean Price
-
Collateral
-
CSA
-
CVA
-
DVA
-
FVA
-
MVA
-
KVA
-
Funding
-
Initial Margin
-
All-in pricing
-
Neural XVA approximators
RiskLab 14 — Clean Price to AI All-In Price
Starting from an Interest Rate Swap:
Clean Price + CVA + DVA + FVA + MVA → Dealer All-In Price
Compare Monte Carlo with an AI surrogate.
Module 15 — Initial Margin, SIMM & Collateral Optimisation with AI The Hidden Cost of Derivatives
​
This module integrates margin, collateral and funding into pricing and portfolio optimisation.
Key Topics
-
Initial Margin
-
Variation Margin
-
SIMM
-
Sensitivities
-
Margin forecasting
-
Collateral eligibility
-
Haircuts
-
Liquidity cost
-
Funding cost
-
MVA
-
Collateral optimisation
-
AI optimisation
RiskLab 15 — AI Initial Margin & Collateral Optimiser
Optimise collateral allocation to minimise:
Funding Cost + Haircuts + Liquidity Cost + MVA
subject to eligibility constraints.
Module 16 — Generative AI for Derivatives Pricing & Model Intelligence
​
The Quant Copilot
Generative AI is used to interpret contracts, retrieve model documentation, select pricing approaches and support quantitative workflows.
Key Topics
-
Large Language Models
-
Retrieval-Augmented Generation
-
Term-sheet interpretation
-
Contract extraction
-
Payoff recognition
-
Model recommendation
-
Documentation
-
Code generation
-
Testing
-
Model limitations
-
Hallucination control
-
Human approval
RiskLab 16 — Derivatives Pricing Copilot
The system receives a term sheet and extracts:
Underlying | Payoff | Dates | Barriers | Optionality | Currency
It then proposes:
Pricing Model + Required Market Data + Calibration Method + Risk Factors
Module 17 — Autonomous AI Agents for the Derivatives Desk
​
From Pricing Engine to Agentic Workflow
This module develops multi-agent architectures capable of coordinating the complete derivatives valuation process.
Key Topics
-
Agentic AI
-
Autonomous workflows
-
Tool calling
-
Market Data Agent
-
Calibration Agent
-
Pricing Agent
-
Greeks Agent
-
Hedging Agent
-
Validation Agent
-
Model Risk Agent
-
Audit trails
-
Human-in-the-loop
RiskLab 17 — Autonomous AI Derivatives Desk
Build:
Trade Agent
↓
Market Data Agent
↓
Calibration Agent
↓
Pricing Agent
↓
Greeks Agent
↓
Hedging Agent
↓
Validation Agent
The validation layer may reject or escalate results when inconsistencies are detected.
Module 18 — AI Pricing Model Risk, Uncertainty & Validation
A Fast Price Is Useless If You Cannot Trust It
​
This module develops a validation framework specifically for AI-based pricing models.
Key Topics
-
AI Model Risk
-
Surrogate-model validation
-
Benchmark models
-
Interpolation risk
-
Extrapolation risk
-
Distribution shift
-
Out-of-distribution detection
-
Conformal Prediction
-
Prediction intervals
-
Stability
-
Explainability
-
Model monitoring
RiskLab 18 — AI Pricing Model Validation Lab
​
Introduce a derivative outside the training domain.
The model must determine whether to:
Price with AI
or
Fallback to Full Numerical Model
Compare:
AI Price | Benchmark Price | Uncertainty Interval | Pricing Error
Module 19 — Quantum Accelerated Pricing & Quantum Monte Carlo
​
Beyond Classical Simulation
This module introduces Quantum Amplitude Estimation and Quantum Monte Carlo as potential alternatives to classical simulation for computationally intensive pricing problems.
Key Topics
-
Classical Monte Carlo
-
Quantum Amplitude Estimation
-
Quantum Monte Carlo
-
State preparation
-
Amplitude encoding
-
European options
-
Basket options
-
Path-dependent payoffs
-
Quantum speed-up
-
Resource requirements
-
Quantum error
-
Classical vs Quantum benchmarking
RiskLab 19 — Monte Carlo vs Quantum Amplitude Estimation
Price the same derivative using:
Classical Monte Carlo
vs
Quantum Amplitude Estimation
and compare:
Accuracy | Number of Evaluations | Circuit Resources | Theoretical Speed-Up
Module 20 — Hybrid Quantum-AI Derivatives Pricing & Hedging
​
The Future Computational Stack
The final module combines classical quantitative finance, Artificial Intelligence and Quantum Computing into a hybrid derivatives architecture.
Key Topics
-
Hybrid quantum-classical workflows
-
Quantum optimisation
-
QAOA
-
Variational quantum algorithms
-
Quantum kernels
-
Quantum Neural Networks
-
Quantum calibration
-
Quantum portfolio optimisation
-
Quantum hedging
-
NISQ constraints
-
Hardware limitations
-
Future fault-tolerant architectures
RiskLab 20 — Hybrid Quantum-AI Hedging Optimiser
An AI engine generates prices and Greeks.
The hedge is then formulated as:
Minimise
Residual Risk + Transaction Costs
subject to:
Delta + Gamma + Vega Constraints
Compare:
Classical Optimisation
vs
Genetic Algorithm
vs
QAOA Quantum Optimisation
The Complete Architecture
MARKET DATA
↓
AI DATA INTELLIGENCE
↓
CURVES & VOLATILITY
↓
AI CALIBRATION
↓
PRICING
↓
GREEKS
↓
PORTFOLIO REVALUATION
↓
DEEP HEDGING
↓
ALL-IN PRICE
↓
MODEL VALIDATION
↓
GENERATIVE AI & AUTONOMOUS AGENTS
↓
QUANTUM ACCELERATION
