AI in Finance Course

Lesson 1 • Data Fundamentals for Finance AI

Explains the types of financial data and their quality requirements. Grounds students in why data preparation is critical before any modeling.

Lesson 2 • History of AI in Financial Services

Traces AI adoption from early rule-based systems to modern neural networks. Provides context for understanding current capabilities and limitations.

Lesson 3 • What AI Means for Finance

Defines AI, machine learning, and deep learning in plain terms. Establishes why finance is a high-value domain for AI adoption.

Lesson 4 • AI Ecosystem and Key Stakeholders

Maps the roles of data scientists, engineers, business analysts, and regulators in AI projects. Clarifies how finance professionals fit into cross-functional AI teams.

Lesson 1 • Regression Models in Finance

Covers linear and nonlinear regression for forecasting prices, returns, and risk metrics. Demonstrates how model assumptions affect financial predictions.

Lesson 2 • Model Evaluation and Selection

Teaches rigorous model comparison using financial performance metrics alongside statistical ones. Prevents overfitting through cross-validation and walk-forward testing.

Lesson 3 • Supervised Learning Workflow

Walks through the end-to-end process of training a supervised model on financial data. Establishes a repeatable workflow used throughout the chapter.

Lesson 4 • Classification Models for Finance

Applies logistic regression, decision trees, and ensemble methods to binary finance outcomes. Connects classification outputs to credit and fraud decisions.

Lesson 1 • Traditional Versus AI Scorecards

Contrasts logistic regression scorecards with gradient boosting and neural network alternatives. Highlights trade-offs between accuracy and regulatory explainability.

Lesson 2 • Credit Model Monitoring and Drift

Establishes processes for detecting population shift and model degradation over time. Ensures credit models remain accurate and fair after deployment.

Lesson 3 • Model Explainability for Credit

Applies SHAP values and LIME to generate explanations for individual credit decisions. Meets regulatory requirements for adverse action notices.

Lesson 4 • Credit Risk Fundamentals

Reviews probability of default, loss given default, and exposure at default as modeling targets. Anchors AI methods in the credit risk framework used by lenders.

Lesson 1 • Graph Analytics for Financial Crime

Uses network analysis to uncover money laundering rings and collusive fraud patterns. Extends detection beyond individual transactions to entity relationships.

Lesson 2 • Supervised Fraud Classification

Applies oversampling, cost-sensitive learning, and ensemble models to labeled fraud datasets. Optimizes for precision-recall trade-offs relevant to fraud operations teams.

Lesson 3 • Fraud Landscape and Data Challenges

Characterizes fraud types and the severe class imbalance that makes detection difficult. Sets the stage for specialized sampling and modeling strategies.

Lesson 4 • Anomaly Detection Methods

Covers isolation forests, autoencoders, and statistical outlier methods for unsupervised fraud detection. Connects anomaly scores to operational alert thresholds.

Lesson 1 • Quantitative Trading Foundations

Introduces alpha generation, signal construction, and the systematic trading lifecycle. Provides the quantitative finance vocabulary needed for AI strategy development.

Lesson 2 • ML-Based Signal Generation

Applies classification and regression models to generate buy, sell, and hold signals from market data. Addresses look-ahead bias and data leakage in financial ML.

Lesson 3 • Reinforcement Learning for Trading

Frames trading as a sequential decision problem and trains agents using reward-based learning. Covers environment design, reward shaping, and policy evaluation.

Lesson 4 • Portfolio Optimization with AI

Extends mean-variance optimization with ML-estimated inputs and deep learning allocation models. Balances return maximization against risk constraints in dynamic markets.

Lesson 1 • NLP Pipeline Deployment

Packages NLP models into production pipelines that process financial text at scale. Covers API design, latency management, and model versioning.

Lesson 2 • Financial Text Data Sources

Catalogs earnings calls, filings, news, and social media as NLP inputs. Explains how text data complements quantitative signals in finance.

Lesson 3 • Large Language Models in Finance

Explores how large language models support financial research, summarization, and Q&A. Addresses hallucination risks and grounding strategies for financial accuracy.

Lesson 4 • Information Extraction from Documents

Applies named entity recognition and relation extraction to financial filings and contracts. Automates extraction of key financial figures and risk disclosures.

Lesson 1 • Stress Testing and Scenario Analysis

Uses generative models and adversarial scenarios to stress-test portfolios under extreme conditions. Satisfies regulatory stress testing requirements with AI-generated scenarios.

Lesson 2 • Liquidity Risk and AI

Forecasts cash flow shortfalls and liquidity gaps using ML on transactional and market data. Enables proactive liquidity management rather than reactive responses.

Lesson 3 • Market Risk Measurement with AI

Applies ML to improve Value at Risk and Expected Shortfall estimation beyond historical simulation. Captures nonlinear risk exposures that traditional models miss.

Lesson 4 • Operational Risk and Anomaly Detection

Detects process failures, cyber threats, and rogue behavior using unsupervised and supervised AI. Reduces operational losses through earlier detection and automated escalation.

Lesson 1 • Bias Detection and Mitigation

Measures disparate impact and demographic parity in credit, insurance, and hiring AI models. Applies pre-processing, in-processing, and post-processing bias remediation techniques.

Lesson 2 • AI Governance Framework Design

Constructs an enterprise AI governance structure covering policies, roles, and review boards. Operationalizes responsible AI from strategy through production monitoring.

Lesson 3 • Explainability Frameworks

Compares global and local explainability methods for communicating AI decisions to stakeholders. Selects appropriate methods based on audience and regulatory context.

Lesson 4 • Regulatory Compliance for AI

Maps AI governance requirements from financial regulators and data protection authorities. Builds compliance workflows into the AI development lifecycle.