Colin de la Higuera
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Colin de la Higuera has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- James Scicluna2 shared publications
- Rémi Eyraud
Qarma Team, LIF Marseille, France
2 shared publications - Achilles A. Beros
Department of Mathematics, University of Hawaii at Manoa, Honolulu, USA
1 shared publication - Alexis Linard
Institute for Computing and Information Science Radboud University Nijmegen The Netherlands
1 shared publication - Barbara Class
Université de Genève
1 shared publication - Christine Solnon
INSA Lyon, CITI, INRIA CHROMA, Villeurbanne, France
1 shared publication
Latest Publications
- Open Access
- 19 July 2024
- Published by Cantonal and University Library Fribourg in Journal of Ethics in Higher Education
- No. 4,p. 231-244
From Disposable Education to Acting in the World as a Human in the Time of AI
This contribution presents two perspectives: one based on AI expertise and the other on the importance of knowledge as human knowledge. The later focuses on the purpose of education by revisiting Alexander von Humboldt’s concept of Bildung as the main purpose of education. Resisting the shifting to radically pragmatic models of education, without Bildung, in recent decades, in the Global North which develops a system of disposable education that focuses on training a workforce for the market. Concerned by the production of disposable knowledge and assessments, with the sole goal of achieving good grades and securing employment, we see Generative AI, as mirroring this functionalist and economist vision of society, revealing strongly the limits of this paradigm. This article first explains how automatic learning, the foundation of generative AI, works. Second, we shall discuss human deep learning, as opposed to surface learning showing how it is required to develop as a person in the Bildung perspective. Third, we highlight this first intuition that AI reveals the functioning of our societies and move beyond, with an art of questioning in philosophy.
- 19 July 2024
- Published by Cantonal and University Library Fribourg in Journal of Ethics in Higher Education
- No. 4,p. 231-244
Open AccessFrom Disposable Education to Acting in the World as a Human in the Time of AI
This contribution presents two perspectives: one based on AI expertise and the other on the importance of knowledge as human knowledge. The later focuses on the purpose of education by revisiting Alexander von Humboldt’s concept of Bildung as the main purpose of education. Resisting the shifting to radically pragmatic models of education, without Bildung, in recent decades, in the Global North which develops a system of disposable education that focuses on training a workforce for the market. Concerned by the production of disposable knowledge and assessments, with the sole goal of achieving good grades and securing employment, we see Generative AI, as mirroring this functionalist and economist vision of society, revealing strongly the limits of this paradigm. This article first explains how automatic learning, the foundation of generative AI, works. Second, we shall discuss human deep learning, as opposed to surface learning showing how it is required to develop as a person in the Bildung perspective. Third, we highlight this first intuition that AI reveals the functioning of our societies and move beyond, with an art of questioning in philosophy.
- 3 June 2022
- Published by Springer Nature
- p. 333-347
Ethical Principles in Industry 4.0 Automated Decision-Making Systems
- 3 June 2022
- Published by Springer Nature
- p. 333-347
Ethical Principles in Industry 4.0 Automated Decision-Making Systems
- Open Access
Learning Unions of k-Testable Languages
A classical problem in grammatical inference is to identify a language from a set of examples. In this paper, we address the problem of identifying a union of languages from examples that belong to several different unknown languages. Indeed, decomposing a language into smaller pieces that are easier to represent should make learning easier than aiming for a too generalized language. In particular, we consider k-testable languages in the strict sense (k-TSS). These are defined by a set of allowed prefixes, infixes (sub-strings) and suffixes that words in the language may contain. We establish a Galois connection between the lattice of all languages over alphabet {\Sigma}, and the lattice of k-TSS languages over {\Sigma}. We also define a simple metric on k-TSS languages. The Galois connection and the metric allow us to derive an efficient algorithm to learn the union of k-TSS languages. We evaluate our algorithm on an industrial dataset and thus demonstrate the relevance of our approach.
Open AccessLearning Unions of k-Testable Languages
A classical problem in grammatical inference is to identify a language from a set of examples. In this paper, we address the problem of identifying a union of languages from examples that belong to several different unknown languages. Indeed, decomposing a language into smaller pieces that are easier to represent should make learning easier than aiming for a too generalized language. In particular, we consider k-testable languages in the strict sense (k-TSS). These are defined by a set of allowed prefixes, infixes (sub-strings) and suffixes that words in the language may contain. We establish a Galois connection between the lattice of all languages over alphabet {\Sigma}, and the lattice of k-TSS languages over {\Sigma}. We also define a simple metric on k-TSS languages. The Galois connection and the metric allow us to derive an efficient algorithm to learn the union of k-TSS languages. We evaluate our algorithm on an industrial dataset and thus demonstrate the relevance of our approach.
- 30 September 2016
- Published by SAGE Publications in Fundamenta Informaticae
- Vol. 146 (4) , 379-402
Grammatical Inference of PCFGs Applied to Language Modelling and Unsupervised Parsing
Recently, different theoretical learning results have been found for a variety of context-free grammar subclasses through the use of distributional learning [1]. However, these results are still not extended to probabilistic grammars. In this work, w
- 30 September 2016
- Published by SAGE Publications in Fundamenta Informaticae
- Vol. 146 (4) , 379-402
Grammatical Inference of PCFGs Applied to Language Modelling and Unsupervised Parsing
Recently, different theoretical learning results have been found for a variety of context-free grammar subclasses through the use of distributional learning [1]. However, these results are still not extended to probabilistic grammars. In this work, w
- 30 September 2016
- Published by SAGE Publications in Fundamenta Informaticae
- Vol. 146 (4) , 431-459
A Canonical Semi-Deterministic Transducer
We prove the existence of a canonical form for semi-deterministic transducers with sets of pairwise incomparable output strings. Based on this, we develop an algorithm which learns semi-deterministic transducers given access to translation queries. W
- 30 September 2016
- Published by SAGE Publications in Fundamenta Informaticae
- Vol. 146 (4) , 431-459
A Canonical Semi-Deterministic Transducer
We prove the existence of a canonical form for semi-deterministic transducers with sets of pairwise incomparable output strings. Based on this, we develop an algorithm which learns semi-deterministic transducers given access to translation queries. W
- 30 September 2016
- Published by SAGE Publications in Fundamenta Informaticae
- Vol. 146 (4)
Preface
International audienceGrammatical Inference addresses the question of learning finite state machines and formal grammars given information about language. It addresses both theoretical and empirical learning problems, even if this special issue will describe work coming from the more theoretical perspective. Thus we hope this special issue is of interest to the readership of Fundamenta Informaticae
- 30 September 2016
- Published by SAGE Publications in Fundamenta Informaticae
- Vol. 146 (4)
Preface
International audienceGrammatical Inference addresses the question of learning finite state machines and formal grammars given information about language. It addresses both theoretical and empirical learning problems, even if this special issue will describe work coming from the more theoretical perspective. Thus we hope this special issue is of interest to the readership of Fundamenta Informaticae
- 6 July 2016
- Published by Springer Nature
- p. 39-50
Computing the Expected Edit Distance from a String to a PFA
- 6 July 2016
- Published by Springer Nature
- p. 39-50
Computing the Expected Edit Distance from a String to a PFA
- Open Access
On the Computation of Distances for Probabilistic Context-Free Grammars
Probabilistic context-free grammars (PCFGs) are used to define distributions over strings, and are powerful modelling tools in a number of areas, including natural language processing, software engineering, model checking, bio-informatics, and pattern recognition. A common important question is that of comparing the distributions generated or modelled by these grammars: this is done through checking language equivalence and computing distances. Two PCFGs are language equivalent if every string has identical probability with both grammars. This also means that the distance (whichever norm is used) is null. It is known that the language equivalence problem is interreducible with that of multiple ambiguity for context-free grammars, a long-standing open question. In this work, we prove that computing distances corresponds to solving undecidable questions: this is the case for the L1, L2 norm, the variation distance and the Kullback-Leibler divergence. Two more results are less negative: 1. The most probable string can be computed, and, 2. The Chebyshev distance (where the distance between two distributions is the maximum difference of probabilities over all strings) is interreducible with the language equivalence problem.
Open AccessOn the Computation of Distances for Probabilistic Context-Free Grammars
Probabilistic context-free grammars (PCFGs) are used to define distributions over strings, and are powerful modelling tools in a number of areas, including natural language processing, software engineering, model checking, bio-informatics, and pattern recognition. A common important question is that of comparing the distributions generated or modelled by these grammars: this is done through checking language equivalence and computing distances. Two PCFGs are language equivalent if every string has identical probability with both grammars. This also means that the distance (whichever norm is used) is null. It is known that the language equivalence problem is interreducible with that of multiple ambiguity for context-free grammars, a long-standing open question. In this work, we prove that computing distances corresponds to solving undecidable questions: this is the case for the L1, L2 norm, the variation distance and the Kullback-Leibler divergence. Two more results are less negative: 1. The most probable string can be computed, and, 2. The Chebyshev distance (where the distance between two distributions is the maximum difference of probabilities over all strings) is interreducible with the language equivalence problem.
- 14 December 2013
- Published by Springer Nature in Machine Learning
- Vol. 96 (1-2) , 1-3
Introduction to the Special Issue on Grammatical Inference
- 14 December 2013
- Published by Springer Nature in Machine Learning
- Vol. 96 (1-2) , 1-3
Introduction to the Special Issue on Grammatical Inference