Radhia Cousot

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search
Radhia Cousot
RadhiaCousot--Picture.jpg
Born (1947-08-06)6 August 1947
Sakiet Sidi Youssef, Tunisia
Died 1 May 2014(2014-05-01) (aged 67)
New York City
Residence New York City
Nationality French
Alma mater Institut National Polytechnique de Lorraine
Known for Abstract interpretation
Spouse(s) Patrick Cousot
Awards

ACM SIGPLAN Programming Languages Achievement Award

IEEE Computer Society Harlan D. Mills Award
Scientific career
Fields Computer science
Thesis Fondements des méthodes de preuve d'invariance et de fatalité de programmes parallèles (1985)
Doctoral advisor Claude Pair

Radhia Cousot (6 August 1947 – 1 May 2014)[1] was a French computer scientist known for inventing abstract interpretation.

Studies[edit]

Radhia Cousot was born on 6 August 1947, in Sakiet Sidi Youssef in Tunisia, where she survived the massacre of the children in her school on February 8, 1958. She then went to the Lycée de jeunes filles at Sousse, the Lycée français at Algiers and then the Polytechnic School of Algiers (where she was ranked 1st and the only woman). She specialized in mathematical optimization and integer linear programming. Supported by a UNESCO fellowship (1972–1975), she obtained a master's degree in Computer Science (Diplôme d'études approfondies (DEA)) at the Joseph Fourier University of Grenoble in 1972. She obtained her Doctorate ès Sciences/State Doctorate in Mathematics in Nancy in 1985 under the supervision of Claude Pair (fr).[nb 1]

Career[edit]

Radhia Cousot was appointed Associate research scientist at the IMAG laboratory of the Joseph Fourier University of Grenoble (1975–1979) and, from 1980 on, at the Centre national de la recherche scientifique, as junior research scientist, research scientist, senior research scientist, and senior research scientist emerita at the Computer Science laboratories of the Henri Poincaré University of Nancy (1980–1983), the University of Paris-Sud at Orsay (1984–1988), the École Polytechnique (1989–2008) where from 1991 she headed the research team “Semantics, Proof and Abstract interpretation”, and the École Normale Supérieure (2006–2014).

Scientific achievements[edit]

Together with her husband Patrick, Radhia Cousot is the originator of abstract interpretation,[2][3] an influential technique in formal methods. Abstract interpretation is based on three main ideas.

  1. Any reasoning/proof/static analysis on a computer system refers to a semantics describing, at some level of abstraction, its possible executions.
  2. The reasoning/proof/static analysis should abstract away all semantic properties irrelevant to the reasoning.
  3. Because of undecidability, sound, fully automated, and always terminating reasonings on/proofs/static analysis of computer systems must perform mathematical inductions in the abstract and so, can only be approximate (even with finiteness and decidability hypothesis, because of combinatorial explosion beyond tiny systems).

In her thesis, Radhia Cousot advanced the semantics, proof, and static analysis methods for concurrent and parallel programs.[4]

Radhia Cousot is at the origin of the contacts with Airbus in January 1999 that lead to the development of Astrée run-time error analyzer from 2001 onwards, a tool for sound static program analysis of embedded control/command software developed at the École Normale Supérieure[5] and now distributed by AbsInt GmbH,[6] a German software company specialized on static analysis. Astrée is used in the transportation, space, and medical software industries.

Awards[edit]

With Patrick Cousot, she received the ACM SIGPLAN Programming Languages Achievement Award [7] in 2013 and the IEEE Computer Society IEEE Computer Society Harlan D. Mills award [8] in 2014 for “the invention of ‘abstract interpretation’, development of tool support, and its practical application”.

Radhia Cousot best young researcher paper award[edit]

Since September 2014, the Radhia Cousot best young researcher paper award[9] is attributed annually by the program chair on behalf of the program committee of the Static Analysis Symposia (SAS).[10]

  • 2014 (Munich, Germany): Aleksandar Chakarov (University of Colorado, Boulder, CO, USA), Expectation invariants for probabilistic program loops as fixed points (with Sriram Sankaranarayanan), M. Müller-Olm & H. Seidl (Eds.): SAS 2014, LNCS 8723, pp. 85–100, Springer
  • 2015 (Saint Malo, France): Marianna Rapoport (University of Waterloo, Ontario, Canada), Precise Data Flow Analysis in the Presence of Correlated Method Calls, (with Ondrej Lhoták and Frank Tip), S. Blazy & T. Jensen (Eds.): SAS 2015, LNCS 9291, pp. 54–71, Springer
  • 2016 (Edinburgh, Scotland): Stefan Schulze Frielinghaus (Technische Universität München, Germany), Enforcing Termination of Interprocedural Analysis, (with Helmut Seidl and Ralf Vogler), Xavier Rival (Ed.): SAS 2016, LNCS 9837, pp. 447–468, Springer
  • 2017 (New York, NY, USA): Suvam Mukherjee (Indian Institute of Science, Bangalore, India) and Oded Padon (Tel Aviv University, Israel), Thread-Local Semantics and its Efficient Sequential Abstractions for Race-Free Programs, (with Sharon Shoham, Deepak D'Souza, and Noam Rinetzky), Francesco Ranzato (Ed.): SAS 2017, LNCS 10422, pp 253–276, Springer

Notes[edit]

  1. ^ In the 1980s, there existed in France two levels of PhDs, the higher one, the Doctorate ès Sciences/State Doctorate being necessary to access professorships. It has since been replaced by the habilitation.

References[edit]

External links[edit]