We propose another approach to type inference with first-class implicit polymorphism, based on the interleaving of an Algorithm $\mathcal M$-style constraint-generating elaboration of terms and a solver for the generated constraints. The novelty of our approach is that types in generated constraints include a explicit syntactic representation of instantiations of unknown polymorphic types. Solving unification constraints with unknown polymorphic types also computes type arguments for the elaborated terms. The resulting system is uniform, simple, and compares favorably with other approaches to first-class implicit polymorphism.

Tue 25 Aug

Displayed time zone: Eastern Time (US & Canada) change

15:30 - 17:00
Types, Semantics, and Probabilistic ProgrammingICFP Papers at IP126 Auditorium
Chair(s): Leonidas Lampropoulos University of Maryland at College Park
15:30
18m
Talk
Another Type Inference Algorithm for First-class Implicit Polymorphism
ICFP Papers
J. Garrett Morris University of Iowa
DOI
15:48
18m
Talk
Same Coeffect, Different Base: Connecting Two Dominant Approaches to Graded Types
ICFP Papers
Vilem-Benjamin Liepelt University of Kent, UK, Danielle Marshall University of Glasgow, Dominic Orchard University of Cambridge; University of Kent
DOI
16:06
18m
Talk
Towards a Higher-Order Bialgebraic Denotational Semantics
ICFP Papers
Sergey Goncharov University of Birmingham, Marco Peressotti University of Southern Denmark, Stelios Tsampas University of Southern Denmark, Henning Urbat University of Erlangen-Nuremberg, Stefano Volpe University of Southern Denmark
DOI
16:24
18m
Talk
LazyHMC: Hamiltonian Monte Carlo simulation for lazy, infinite dimensional probabilistic programs
ICFP Papers
Maria-Nicoleta Craciun University of Oxford, C.-H. Luke Ong NTU, Tom Schrijvers KU Leuven, Sam Staton University of Oxford
DOI
16:42
18m
Talk
Imprecise Probabilistic Programming, Precisely (Functional Pearl)
ICFP Papers
Jack Liell-Cock University of Oxford, Sam Staton University of Oxford
DOI