Supervisor: Frédéric Blanqui and Floris van Doorn
Place: Deducteam, Laboratoire Méthodes Formelles (LMF), ENS Paris-Saclay, 4 avenue des Sciences, Gif-sur-Yvette, and Mathematics Department, University Paris-Saclay, bâtiment 307, rue Michel Magat, Orsay (there are 2.7 km between the two places)
Context: This project is about formal proofs as digital objects, more specifically the translation of formal proofs between different proof systems. Formal proofs are used in mathematics but also in the industry for certifying the correctness of protocols, software and hardware.
Interoperability is a very important feature in computer science and engineering to avoid useless work duplication and allow more safety. Unfortunately, interoperability between proof systems is not well developed. One important difficulty is that proof systems may have incompatible features: their combination may be inconsistent. Therefore, to translate a proof from one system to the other, we need to analyze the features of the first system used in the proof and check whether they are compatible with the features of the target system.
The λΠ-calculus modulo rewriting, and its implementation Dedukti, is a powerful logical framework allowing users to define their own logic and represent the proofs in those logics [1,2]. For instance, one can represent in Dedukti first-order logic and its proofs, simple type theory and its proofs, the Isabelle logic and its proofs [5], the Coq logic and its proofs [3], the Agda logic and its proofs [6], etc. In addition, there is a number of tools for transforming those proofs and translate them back to various other systems: HOL-Light, Coq, PVS, Lean, etc. [7].
There exist several tools for checking the correctness of Dedukti files: dkcheck, kontroli and lambdapi. While dkcheck and kontroli are mere checkers taking complete Dedukti files as input, Lambdapi is a proof assistant featuring implicit arguments, type inference, coercions, tactics, the possibility of calling external automated theorem provers, etc. for building Dedukti proofs interactively.
Lean4[8] is a dependently typed proof assistant and programming language. It has an expressive logic with inductive types and universes, similar to Coq. It has a powerful elaborator, tactic language, macro expansion system [9] and compiler. It's predecessor Lean3 has a large mathematical library mathlib, that will be ported to Lean4 in 2022, and is of great interest for other proof systems.
Objective: The goal of this project is to develop in Lean4 a translator from Lean4 to Dedukti. Compared to the encodings already done for Coq or Agda, a possible difficulty is the handling of proof irrelevance, which is a specific feature of Lean meaning that any two proofs of the same proposition are identified [10].
Workplan: One will start by getting some familiarity with the λΠ-calculus modulo rewriting and the Dedukti or Lambdapi languages, the meta-theory of Lean [10], and the encodings of Coq and Agda in Dedukti [3,6]. Concerning proof irrelevance, the intern could perhaps get some inspiration from the encoding of a particular case of proof irrelevance in [4].
Requirements: Some familiarity with Lean.
References:
[1] Dedukti: a logical framework based on the λΠ-calculus modulo theory, A. Assaf, G. Burel, R. Cauderlier, D. Delahaye, G. Dowek, C. Dubois, F. Gilbert, P. Halmagrand, O. Hermant, and R. Saillard, Draft, 2016.
[2] Some axioms for mathematics, F. Blanqui, G. Dowek, E. Grienenberger, G. Hondet and F. Thiré, FSCD 2021.
[3] Higher-Order Confluence and Universe Embedding in the Logical Framework, Gaspard Férey, PhD, 2021.
[4] Encoding of Predicate Subtyping with Proof Irrelevance in the λΠ-Calculus Modulo Theory, Gabriel Hondet and Frédéric Blanqui, TYPES'20.
[5] Translating proofs between Isabelle and Dedukti, Yann Leray, M1 Internship Report, 2021.
[6] Encoding Agda Programs Using Rewriting, G. Genestier, FSCD'20.
[7] Sharing a Library between Proof Assistants: Reaching out to the HOL Family, François Thiré, LFMTP'18.
[8] The Lean 4 Theorem Prover and Programming Language, Leonardo de Moura and Sebastian Ullrich, CADE'21.
[9] Beyond Notations: Hygienic Macro Expansion for Theorem Proving Languages, Sebastian Ullrich and Leonardo de Moura, IJCAR'20.
[10] The type theory of Lean, Mario Carneiro, 2019.