collaborators

6 papers

physics.flu-dyn2026

Gradient-free learning of a closed-loop wall controller for turbulent drag reduction

Giorgio Maria Cavallazzi, Miguel Pérez Cuadrado, Alfredo Pinelli

Closed-loop wall controllers learnt by multi-agent reinforcement learning are usually trained on periodic boxes far smaller than the flows they are meant to drive, and a large part…

physics.flu-dyn2026

Manifold-adapted radial basis functions for reduced-order modelling of chaotic flows

Miguel Pérez Cuadrado, Giorgio Maria Cavallazzi, Alfredo Pinelli

Chaotic systems often evolve on a low-dimensional attractor whose geometry varies from one region to another. We propose a non-intrusive reduced-order model that reads this local g…

physics.flu-dyn2026

Reward hacking in physical reinforcement learning revealed by turbulent drag reduction

Giorgio Maria Cavallazzi, Miguel Pérez-Cuadrado, Alfredo Pinelli

Reinforcement-learning controllers optimise specified rewards, but in physical systems those rewards often capture only part of the true control objective. Three mechanisms through…

physics.flu-dyn2026

Deep reinforcement learning with spatial and temporal awareness for active boundary control of buoyancy-driven convection

Giorgio Maria Cavallazzi, Miguel Pérez Cuadrado, Alfredo Pinelli

Deep reinforcement learning (DRL) applied to thermal convection control consistently produces degenerate actuation: wall-temperature policies whose outputs are saturated, pseudo-ra…

math.NA2026

Restoring Convergence Order in Explicit Runge-Kutta Integration of Hyperbolic PDE with Time-Dependent Boundary Conditions

Giorgio Maria Cavallazzi, Miguel Pérez Cuadrado, Alfredo Pinelli

Explicit Runge-Kutta (RK) integration of hyperbolic initial-boundary value problems with time-dependent Dirichlet data often displays order reduction: the observed convergence orde…

physics.comp-ph2025

Walsh-Hadamard Neural Operators for Solving PDEs with Discontinuous Coefficients

Giorgio M. Cavallazzi, Miguel Pérez Cuadrado, Alfredo Pinelli

Neural operators have emerged as powerful tools for learning solution operators of partial differential equations (PDEs). However, standard spectral methods based on Fourier transf…