Dec 8 – 10, 2025
CERN
Europe/Zurich timezone

Optimal Jacobian Accumulation is NP-hard

Dec 8, 2025, 4:15 PM
20m
31/3-004 - IT Amphitheatre (CERN)

31/3-004 - IT Amphitheatre

CERN

105
Show room on map
Contributed Talk Contributed Talks

Speaker

Blair D. Sullivan

Description

Resolving a long-standing open question, we show that that the core AD problem of accumulating a Jacobian matrix while minimizing multiplications is NP-complete. Complementing this, we show that the running time of a relatively straight-forward $O^*(2^n)$ algorithm is essentially best possible under the Exponential Time Hypothesis. We also establish NP-completeness for the 'scarcity' problem of obtaining a minimum-size matrix-free Jacobian representation.

Importantly, we formulate both problems graph-theoretically, using sequences of vertex eliminations in a directed acyclic graph. Our results are facilitated by several illustrative structural insights about sequences of vertex eliminations, which also lead to a novel SAT encoding for Jacobian accumulation. We conclude with a discussion of open problems related to parameterized algorithms, approximations, and problem variants allowing edge eliminations. In addition to sharing our results, we hope to instigate more collaboration between graph algorithms researchers and the AD community.

Author

Co-authors

Alex Crane (University of Utah) Matthias Bentert (TU Berlin) Pal Drange (University of Bergen) Yosuke Mizutani

Presentation materials