A note on phase transitions and computational pitfalls of learning from sequences

An ever greater range of applications call for learning from sequences. Grammar induction is one prominent tool for sequence learning, it is therefore important to know its properties and limits. This paper presents a new type of analysis for inductive learning. A few years ago, the discovery of a p...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Journal of intelligent information systems Ročník 31; číslo 2; s. 177 - 189
Hlavní autoři: Cornuéjols, Antoine, Sebag, Michèle
Médium: Journal Article
Jazyk:angličtina
Vydáno: Boston Springer US 01.10.2008
Springer Nature B.V
Témata:
Algorithms
Artificial Intelligence
Bioinformatics
Computer Science
Data Structures and Information Theory
Grammar
Heuristic
Hypotheses
Information Storage and Retrieval
IT in Business
Language
Logic programming
Natural Language Processing (NLP)
Phase transitions
Studies
Supervised learning
Sequence learning
Grammatical inference
ISSN:0925-9902, 1573-7675
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:An ever greater range of applications call for learning from sequences. Grammar induction is one prominent tool for sequence learning, it is therefore important to know its properties and limits. This paper presents a new type of analysis for inductive learning. A few years ago, the discovery of a phase transition phenomenon in inductive logic programming proved that fundamental characteristics of the learning problems may affect the very possibility of learning under very general conditions. We show that, in the case of grammatical inference, while there is no phase transition when considering the whole hypothesis space, there is a much more severe “gap” phenomenon affecting the effective search space of standard grammatical induction algorithms for deterministic finite automata (DFA). Focusing on standard search heuristics, we show that they overcome this difficulty to some extent, but that they are subject to overgeneralization. The paper last suggests some directions to alleviate this problem.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-2
content type line 23
ISSN:0925-9902
1573-7675
DOI:10.1007/s10844-008-0063-6