Pricing options on flow forwards by neural networks in Hilbert space
This addresses a specific problem in financial mathematics for pricing complex derivatives, representing an incremental advance with a novel method for a known bottleneck.
The authors tackled the problem of pricing options on flow forwards by recasting it as an optimization in a Hilbert space and solving it with a novel infinite-dimensional neural network architecture, achieving excellent numerical efficiency and superior performance over classical methods.
We propose a new methodology for pricing options on flow forwards by applying infinite-dimensional neural networks. We recast the pricing problem as an optimization problem in a Hilbert space of real-valued function on the positive real line, which is the state space for the term structure dynamics. This optimization problem is solved by facilitating a novel feedforward neural network architecture designed for approximating continuous functions on the state space. The proposed neural net is built upon the basis of the Hilbert space. We provide an extensive case study that shows excellent numerical efficiency, with superior performance over that of a classical neural net trained on sampling the term structure curves.