iBet uBet web content aggregator. Adding the entire web to your favor.
iBet uBet web content aggregator. Adding the entire web to your favor.



Link to original content: https://www.irit.fr/~Ralph.Matthes/
Homepage of Ralph Matthes

Homepage of Ralph Matthes, Chargé de Recherche au CNRS

email address: irit.fr, preceded by my family name and the @ sign

I have a permanent CNRS research position at IRIT (Institute for Computer Science of Toulouse), located on the campus of the Technical University Toulouse. My postal address is (the first line is essential with both ingredients!)

IRIT - Université Paul Sabatier
Ralph Matthes, Équipe ACADIE
118 route de Narbonne
F-31062 Toulouse Cedex 9
France

I am member of the research group ACADIE that is the common successor to SVF and ZENO. The French acronym stands for "Assistance with the Certification of Distributed and Embedded Applications".

As a CNRS researcher, I belong to Section 6 of the National Committee of Scientific Research (after splitting the former 7 into 6 and 7).

A recent photo: A
 picture of me

Research Interests

I am working in Theoretical Computer Science, but this could as well be called Foundations of Computer Science. Within theoretical computer science, I tend towards logic in computer science, semantics and theory of programming. My work resides at the interface between mathematics and computer science. I am strongly interested in computer formalization of my research results and thus in having automatically verifiable artefacts representing these results. This necessitates (and thus comprises) the computer formalization of mathematical structures, such as data types and their recursion schemes and, in general, the structures known from category theory. Still, for me, computer formalization of known mathematics is not an end in itself and only a prerequisite for formalization projects aiming at establishing new results.

My work is based on the following research topics: lambda calculus, proof theory, term rewriting, type theory, interactive theorem proving, coinductive methods, category theory - the latter in particular as a semantic framework for lambda-calculus, type theory and coinductive methods. Category theory had even long before inspired many notions in my analysis of recursion schemes, but in my current research, category theory is being employed.

My current main research efforts go into two directions:

  1. Representation of variable binding (in particular for lambda calculi) by means of category theory, fully formalized in the UniMath library of univalent mathematics in the Coq theorem prover. This is to say that the syntax of those languages with variable binding is modeled with notions of category theory, as formalized in type theory with dependent types by help of an interactive proof assistant, and concrete instances are then obtained by using the conception of sets according to homotopy type theory - that is provided by the basic building blocks of the UniMath library. Since 2022, the representations I study also include the case where variable binding occurs in non-wellfounded (coinductive) syntax. The work is published in conferences and journals and makes the formalization(s) accessible. As a by-product of this research, the notion of displayed monoidal category was developed that proved to be useful for formalizing the semantics of linear logic.
  2. Representation of aggregates of proofs, currently for propositional logic only. Development of coinductive and inductive term notions that stand for all possible outcomes of proof search. Deriving decision procedures for existence of inhabitants in simply-typed lambda-calculus, but also for the question of finiteness of the number of (normal) inhabitants and related questions. This research is currently not computer formalized but only exists in the classical form of journal and conference papers.

In the past, I directed many of my publications towards uses of parametric polymorphism to guarantee the termination of inductive and coinductive recursion schemes for data types that could come as families and that could go beyond strictly positive data types - the latter made them unfeasible for the Coq theorem prover, but such schemes could still be formally justified by a shallow embedding into Coq (assuming impredicative Set). Coinductive methods played an important role for the representation of graphs and their transformations. Even earlier, I worked just on lambda-calculi and their foundational properties, in particular for extensions and variations that featured permutative/commuting conversions. My accent there was on the careful inductive definition of auxiliary concepts, so as to guarantee smooth proofs following that inductive structure. Permutative/commuting conversions are also a challenge to overcome in my (now old) work on continuation-passing style transformations.

A very important part of the research sketched here was obtained with collaborators. The research interests are mine, and they are being shown here.

Published and unpublished work

European Research Network on Formal Proofs EuroProofNet (formal information by COST association), October 11, 2021 - October 10, 2025 (action chair: Frédéric Blanqui, action vice chair: Sandra Alves).

I am member of the working groups WG3 (Program Verification), WG4 (Libraries of Formal Proofs) and WG6 (Type Theory).

The European research network on types for programming and verification (EUTYPES)

COST Action CA15123 (formal information by COST association), March 21, 2016 - March 20, 2020. I was a member of the management committee throughout the lifetime of the network (for the partner country France).

The TYPES conference series

See the permanent web site. The website of TYPES 2017 also lists the previous instalments, including functioning links to the web sites of 8 of them (before 2017). The TYPES conference series was connected to EUTYPES but not part of it. I was chairman of the steering committee from June 2017 to June 2021 (after TYPES 2017 - TYPES 2021) and had been an ordinary steering committee member before (after TYPES 2013 - TYPES 2017).

Univalent Mathematics

A unified approach to formalization of mathematical knowledge based on Univalent Foundations: the GitHub repository (I have properly contributed to the UniMath library and the additional TypeTheory library, and I am member of the UniMath Coordinating Committee - with six members as of December 2022).

Research project Climt

"Categorical and Logical Methods in Model Transformation", 2012-2016, financed by the French National Research Agency, no. ANR-11-BS02-016. I was the responsible for the Toulouse site (the other site was Grenoble, with project leader Rachid Echahed).

Talks

Teaching (in French in France 2012-presently)

Teaching (in German in Germany 1997-2005)

Events since 2019

12th Workshop Fixed Points in Computer Science 2024 (FICS 2024) on February 19 and 20 in Naples, Italy, a satellite of CSL 2024. I chaired the steering committee.

29th International Conference on Types for Proofs and Programs (TYPES 2023), 12-17 June 2023 in València, Spain. I was in the PC.

11th Workshop Fixed Points in Computer Science 2023 (FICS 2023) on February 17 in Warsaw, Poland, a satellite of CSL 2023. I chaired the steering committee and I was in the PC.

7th International Conference on Formal Structures for Computation and Deduction (FSCD 2022), August 2-8, 2022 in Haifa, Israel. I was in the PC.

School on Univalent Mathematics, July 17-23, 2022 in Cortona, Italy. I was one of the lecturers.

27th International Conference on Types for Proofs and Programs (TYPES 2021), 14 - 18 June 2021 virtually (online), hosted by Leiden University, The Netherlands. I was chair of the steering committee and I in the PC.

School and Workshop on Univalent Mathematics, July 27-31, 2020 in Cortona, Italy (cancelled because of the COVID-19 sanitary crisis). I would have been one of the lecturers/mentors.

26th International Conference on Types for Proofs and Programs (TYPES 2020), 2 - 5 March 2020 in Torino, Italy (cancelled only days before the conference because of the COVID-19 outbreak in Piedmont, with the whole programme already set up). I have chaired the steering committee, and I was in the PC.

18th Workshop Proof, Computation, Complexity 2019 (PCC 2019), July 15-19, 2019 in Djursholm close to Stockholm, Sweden at the Institut Mittag-Leffler. I was in the SC and co-organizer.

International Workshop Logical Frameworks and Meta-Languages: Theory and Practice (LFMTP'19), June 22, 2019 in Vancouver, Canada (affiliated with LICS 2019). I was in the PC.

25th International Conference on Types for Proofs and Programs (TYPES 2019), 11 - 14 June 2019 in Oslo, Norway. I have chaired the steering committee, and I was in the PC.

School and Workshop on Univalent Mathematics, April 1-5, 2019 in Birmingham, UK. I was one of the lecturers/mentors.

To see also older events, please refer to the full list.


Former affiliation until 31st August 2005: Wissenschaftlicher Assistent (C1) at the chair for theoretical computer science, Ludwig-Maximilian-University, Munich
Birthdays of mathematicians
Ralph Matthes
https://www.irit.fr/~Ralph.Matthes/

Last modified (date in French): mer. août 28 17:04:21 CEST 2024