Add AADC driver for automatic training label generation - #1
Open
mar20tinha-dot wants to merge 1 commit into
Open
Add AADC driver for automatic training label generation#1mar20tinha-dot wants to merge 1 commit into
mar20tinha-dot wants to merge 1 commit into
Conversation
Demonstrates AADC (automatic adjoint differentiation) for generating DiffML training labels (prices + pathwise derivatives) automatically. Results on 5 tests: - BS European: exact match with manual pathwise (error < 1e-15) - Heston call: delta AAD/FD = 0.9997, vega AAD/FD = 1.0004 One reverse pass gives both. No manual chain rule derivation. - Heston + barrier: honest assessment — pathwise biased at barrier (confirms Glasserman's point about LRM being needed) - Basket d=20: 25x speedup (20 deltas in one pass vs 40 FD bumps) - Label generation: 3-4x speedup vs FD for Heston 50-step AADC value: generate exact derivatives for arbitrary models without manual formulas. Scales to stochastic vol, multi-asset, path-dependent — record simulation on tape, one reverse pass. Requires: aadc (https://matlogica.com/aadc)
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Adds an AADC (Automatic Adjoint Differentiation) driver that generates DiffML training labels (prices + pathwise derivatives) automatically, without manual derivative formulas.
Results
Key points
File
src/driver_diffml_aadc.py— self-contained, 5 tests + benchmarks.Requires: AADC Python package.