Order $p$ quantum Wasserstein distances from couplings
Emily Beatty, Daniel Stilck França
TL;DR
The paper proposes a broad quantum generalisation of the p-Wasserstein distance, $W_p^d$, defined via couplings of pure states and a metric $d$ on the pure-state projective space. By restricting to separable couplings, the authors obtain a robust, order-$p$ distance that recovers trace distance and Nielsen’s complexity in relevant limits, extends to mixed states, and adapts to diverse underlying geometries. They establish key properties (nonnegativity, symmetry, continuity, dual formulation for $p=1$) and explore important instances such as $W_1^H$ and $W_p^1$, as well as complexity-geometry extensions to mixed states. The work uncovers rich behaviour for random quantum states, including phase transitions with subsystem size, and provides operational interpretations for cq-sources, along with hypercontractivity and potential qWGANs based on these distances. Overall, this framework offers a flexible, unifying approach to quantum optimal transport with broad theoretical and practical implications for quantum information, computation, and machine learning.
Abstract
Optimal transport provides a powerful mathematical framework with applications spanning numerous fields. A cornerstone within this domain is the $p$-Wasserstein distance, which serves to quantify the cost of transporting one probability measure to another. While recent attempts have sought to extend this measure to the realm of quantum states, existing definitions often present certain limitations, such as not being faithful. In this work, we present a new definition of quantum Wasserstein distances. This definition, leveraging the coupling method and a metric applicable to pure states, draws inspiration from a property characterising the classical Wasserstein distance - its determination based on its value on point masses. Subject to certain continuity properties, our definition exhibits numerous attributes expected of an optimal quantum rendition of the Wasserstein distance. Notably, our approach seamlessly integrates metrics familiar to quantum information theory, such as the trace distance. Moreover, it provides an organic extension for metrics, like Nielsen's complexity metric, allowing their application to mixed states with a natural operational interpretation. Furthermore, we analyze this metric's attributes in the context of random quantum states and unveil phase transitions concerning the complexity of subsystems of random states.