Holomorphic Cauchy networks reduce memory and speed computation for high-dimensional Brownian tails

Holomorphic Cauchy Networks for Efficient Stationary Tail Computation in High-Dimensional Reflected Brownian Motions

Machine Learning

Summary

Computing probabilities related to complex random motions in many dimensions is very hard and slow. The authors build a new mathematical network called Holomorphic Cauchy Networks (HCNs) that use complex numbers to represent these motions more efficiently. Their approach reduces memory needs and speeds up calculations compared to previous methods. They show it works well up to 50 dimensions, maintaining accuracy while using fewer resources.

What this means in practice

  • For quantitative finance teams: Reduce computational cost and memory for risk calculations involving multi-dimensional reflected Brownian models used in option pricing and portfolio risk.
  • For machine learning engineers: Improve efficiency of high-dimensional probability distribution estimations when modeling stochastic processes with boundary reflections in feature spaces.

Authors

Kehan Wang

Abstract

Stationary tail computation for high-dimensional reflected Brownian motions (RBMs) remains computationally challenging. Building on NN4MGF's shared-coordinate basic-adjoint-relationship (BAR) framework, we develop Holomorphic Cauchy Networks (HCNs). Native complex-valued rational Cauchy subnetworks enforce holomorphicity on pole-free domains, removing sampled Cauchy-Riemann penalties and their input-derivative graphs. Exact origin anchoring and boundary-coordinate exclusion enforce structural identities, while anchored interactions represent dependent transforms. Under holomorphic-log assumptions, we establish simultaneous value and first-derivative approximation for the constrained architectures and a mixed-derivative lower bound for coordinate-only representations under nonzero cross-covariance. Closed-form real-axis derivatives and sequential backpropagation further control memory. On the same NVIDIA H100 NVL, the tested 30D HCN configuration uses approximately tenfold less peak allocated memory and has 9.7-fold lower mean update time than NN4MGF; extrapolation to the respective scheduled update counts gives a projected full-schedule update-time ratio of 58.2. With 64 million sampled base inputs per high-dimensional run, HCN reaches 50 dimensions on a laptop GPU with 6.367% mean relative error over thresholds 1 through 8. Degree-matched comparisons yield lower reported mean relative errors over the primary evaluation ranges, including 1.422% versus 6.684% on the 2D Harrison non-product benchmark. An exploratory 10D correlated example illustrates the interaction extension.