Aarhus University Seal

Talk: Amaryllis: Integrating Dynamic Allocation and Iris-Style Resources into a Distributional Probabilistic Logic by Janine Lohse

Info about event

Time

Monday 16 November 2026,  at 13:00 - 14:00

Location

Nygaard 192

Abstract:

 

There has recently been exciting progress in the realm of probabilistic separation logics. An important subclass of these—including PSL, Lilac, Bluebell, and pcOL—are distributional probabilistic logics (or DPLs, for short), meaning that they provide primitive Hoare-style assertions about probability distributions on the program state, along with powerful modularity principles like independence and conditioning. However, none of these logics support reasoning about dynamically allocated memory (i.e., pointers into a heap). We argue that this is due to a fundamental obstacle: since the shape of memory (and identity of memory locations) may differ under different random outcomes, it is unclear how pointer ownership can be harmonized with probabilistic independence and conditioning.

 

In this talk, I will present Amaryllis, the first DPL to support independence and conditional reasoning while also handling dynamic memory allocation. To overcome the aforementioned obstacle, we developed a new indexed valuation-style model of probabilistic assertions, whereby ownership of a standard resource (e.g., heaps) can be promoted to ownership at the level of distributions in a generic fashion. I will then analyze how this new model poses challenges for the soundness of crucial locality principles, and describe how we overcome those challenges with our novel notion of a probabilistic frame-preserving update that generalizes Iris' frame-preserving update modality.