AFNO: a certifiably arbitrage-free neural operator framework for European call-price surfaces
chaumian · x · 2026-11-13
arXiv paper 2611.12899 introduces AFNO, a certifiably arbitrage-free and universal encoder-decoder neural operator framework that maps RKHS inputs to convex combinations of arbitrage-free European call-price surfaces via an approximate nearest-neighbour problem. The authors prove a nearest arbitrage-free correction of an unconstrained prediction need not exist—so no-arbitrage cannot generally be restored by post-processing—and show empirically that standard NOs like FNO admit parameter configurations violating static no-arbitrage, which can yield negative PnL in downstream trading tasks.
More from Research
- TLAPS-proof LLM serving engine: 516,973 obligations at just 0.18% latency overhead — tianyin_xu · 2026-11-13
- Single-constraint KL minimization reduces to 1-D bisection on the dual — caglar_ee · 2026-11-13
- AI-Simulated Survey Matches Real Human Rankings: Spearman r=0.886 in Head-to-Head Match-ups — soumitrashukla9 · 2026-11-13
- Professor Uses AI Simulation to Pick His Executive Headshot from Six Looks, Biker Jacket Included — soumitrashukla9 · 2026-11-13
- Researcher pushes back on METR's 2x AI acceleration claim, citing 1.4x figure — HaydnBelfield · 2026-11-13
- Mercor launches OWE-Bench: 26 real engineering projects that stump coding agents — ben_burtenshaw · 2026-11-13