hard · Frm Part 2 Current Issues

A bank uses a black-box neural network for credit limit decisions. To provide 'adverse action notices,' it employs a local surrogate model technique. A validator finds that for a single applicant, two different 'post-hoc' explainers provide conflicting reasons for rejection.

This phenomenon highlights which specific risk in XAI validation?

  1. A violation of 'demographic parity,' where the model is rejecting candidates based on group-level features rather than individual risk profiles.
  2. The lack of 'local fidelity,' where the surrogate model fails to accurately map the complex model's behavior in the specific region of that applicant's data.
  3. The 'exclusion fallacy,' where the model has reconstructed a protected attribute through proxies that the explainers are unable to detect.
  4. Overfitting in the training phase, which prevents the explainers from converging on a stable feature importance ranking.

Sign up free to see the explanation and track your rank →

More Frm Part 2 Current Issues practice

KomFi Academy — Stop doomscrolling. Get KomFi.

Build your intelligence, anytime, anywhere.

KomFi Academy is a curated training platform with 48,000+ practice questions, 20,000+ flashcards, on-demand video lectures, podcasts, and 4K slide decks across the topics serious professionals study: GMAT, LSAT, MCAT, Investment Banking, Private Equity (LBOs & PE math), Private Credit, Quantitative Finance, Financial Accounting, Asset- Backed Securities, Volume Profile Analysis, Order Flow Trading, Market Microstructure, Volume Spread Analysis, Elliott Wave Theory, Volume-Price Analysis, and Public Offering Frameworks.

What's inside

Topics

View pricing · Read testimonials