Results 1  10
of
36
Training Tree Transducers
 IN HLTNAACL
, 2004
"... Many probabilistic models for natural language are now written in terms of hierarchical tree structure. Treebased modeling still lacks many of the standard tools taken for granted in (finitestate) stringbased modeling. The theory of tree transducer automata provides a possible framework to ..."
Abstract

Cited by 128 (11 self)
 Add to MetaCart
Many probabilistic models for natural language are now written in terms of hierarchical tree structure. Treebased modeling still lacks many of the standard tools taken for granted in (finitestate) stringbased modeling. The theory of tree transducer automata provides a possible framework to draw on, as it has been worked out in an extensive literature. We motivate the use of tree transducers for natural language and address the training problem for probabilistic treetotree and treetostring transducers.
A Kleene theorem for weighted tree automata
 Theory of Computing Systems
, 2002
"... In this paper we prove Kleene's result for tree series over a commutative and idempotent semiring A (which is not necessarily complete or continuous), i.e., the class of recognizable tree series over A and the class of rational tree series over A are equal. We show the result by direct automata ..."
Abstract

Cited by 21 (8 self)
 Add to MetaCart
In this paper we prove Kleene's result for tree series over a commutative and idempotent semiring A (which is not necessarily complete or continuous), i.e., the class of recognizable tree series over A and the class of rational tree series over A are equal. We show the result by direct automatatheoretic constructions and prove their correctness.
Relating tree series transducers and weighted tree automata
 INT. J. FOUND. COMPUT. SCI
, 2005
"... ..."
Why Synchronous Tree Substitution Grammars?
"... Synchronous tree substitution grammars are a translation model that is used in syntaxbased machine translation. They are investigated in a formal setting and compared to a competitor that is at least as expressive. The competitor is the extended multi bottomup tree transducer, which is the bottom ..."
Abstract

Cited by 10 (6 self)
 Add to MetaCart
Synchronous tree substitution grammars are a translation model that is used in syntaxbased machine translation. They are investigated in a formal setting and compared to a competitor that is at least as expressive. The competitor is the extended multi bottomup tree transducer, which is the bottomup analogue with one essential additional feature. This model has been investigated in theoretical computer science, but seems widely unknown in natural language processing. The two models are compared with respect to standard algorithms (binarization, regular restriction, composition, application). Particular attention is paid to the complexity of the algorithms. 1
Code selection by tree series transducers, in
 Proc. 9th Int. Conf. on Implementation and Application of Automata, Vol. 3317 of LNCS
, 2004
"... Abstract. In this paper we model code selection by tree series transducers. We are given an intermediate representation of some compiler as well as a machine grammar with weights, which reflect the number of machine cycles of the instructions. The derivations of the machine grammar are machine codes ..."
Abstract

Cited by 9 (0 self)
 Add to MetaCart
(Show Context)
Abstract. In this paper we model code selection by tree series transducers. We are given an intermediate representation of some compiler as well as a machine grammar with weights, which reflect the number of machine cycles of the instructions. The derivations of the machine grammar are machine codes. In general, a machine grammar is ambiguous and hence there might exist more than one derivation of an intermediate code. We show how to filter out a cheapest such derivation and thereby perform tree parsing and tree pattern matching using tree series transducers. 1
Compositions of Tree Series Transformations
, 2005
"... Tree series transformations computed by bottomup and topdown tree series transducers are called bottomup and topdown tree series transformations, respectively. (Functional) compositions of such transformations are investigated. It turns out that the class of bottomup tree series transformations ..."
Abstract

Cited by 8 (6 self)
 Add to MetaCart
(Show Context)
Tree series transformations computed by bottomup and topdown tree series transducers are called bottomup and topdown tree series transformations, respectively. (Functional) compositions of such transformations are investigated. It turns out that the class of bottomup tree series transformations over a commutative and complete semiring is closed under leftcomposition with linear bottomup tree series transformations and rightcomposition with boolean deterministic bottomup tree series transformations. Moreover, it is shown that the class of topdown tree series transformations over a commutative and complete semiring is closed under rightcomposition with linear, nondeleting topdown tree series transformations. Finally, the composition of a boolean, deterministic, total topdown tree series transformation with a linear topdown tree series transformation is shown to be a topdown tree series transformation.
Towards Formal Structural Representation of Spoken Language: An Evolving Transformation System (ETS) Approach
, 2005
"... Speech recognition has been a very active area of research over the past twenty years. Despite an evident progress, it is generally agreed by the practitioners of the field that performance of the current speech recognition systems is rather suboptimal and new approaches are needed. The motivation ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
Speech recognition has been a very active area of research over the past twenty years. Despite an evident progress, it is generally agreed by the practitioners of the field that performance of the current speech recognition systems is rather suboptimal and new approaches are needed. The motivation behind the undertaken research is an observation that the notion of representation of objects and concepts that once was considered to be central in the early days of pattern recognition, has been largely marginalised by the advent of statistical approaches. As a consequence of a predominantly statistical approach to speech recognition problem, due to the numeric, feature vectorbased, nature of representation, the classes inductively discovered from real data using decisiontheoretic techniques have little meaning outside the statistical framework. This is because decision surfaces or probability distributions are difficult to analyse linguistically. Because of the later limitation it is doubtful that the gap between speech recognition and linguistic research can be bridged by the numeric representations. This thesis investigates an alternative, structural, approach to spoken language representation and categorisa
Weighted Extended Tree Transducers
, 2010
"... The first systematic treatment of weighted extended tree transducers (wxtt) over countably complete semirings is provided. It is proved that the extension in the lefthand sides of a wxtt can be simulated by the inverse of a linear and nondeleting tree homomorphism. In addition, a characterization o ..."
Abstract

Cited by 4 (4 self)
 Add to MetaCart
The first systematic treatment of weighted extended tree transducers (wxtt) over countably complete semirings is provided. It is proved that the extension in the lefthand sides of a wxtt can be simulated by the inverse of a linear and nondeleting tree homomorphism. In addition, a characterization of weighted tree transformations computed by bottom up wxtt in terms of bimorphisms is provided. Backward and forward application of wxtt to recognizable weighted tree languages are considered. It is shown that the backward application of a linear wxtt preserves recognizability and that the domain of an arbitrary bottomup wxtt is recognizable. Examples demonstrate that neither backward nor forward application of arbitrary wxtt preserves recognizability. Finally, a Hasse diagram relates most of the important subclasses of weighted tree transformations computed by wxtt.
Does osubstitution preserve recognizability?
 IN PROC. 11TH INT. CONF. IMPLEM. AND APPL. OF AUTOMATA
, 2006
"... Substitution operations on tree series are at the basis of systems of equations (over tree series) and tree series transducers. Tree series transducers seem to be an interesting transformation device in syntactic pattern matching. In this contribution, it is shown that osubstitution preserves reco ..."
Abstract

Cited by 4 (3 self)
 Add to MetaCart
Substitution operations on tree series are at the basis of systems of equations (over tree series) and tree series transducers. Tree series transducers seem to be an interesting transformation device in syntactic pattern matching. In this contribution, it is shown that osubstitution preserves recognizable tree series provided that the target tree series is linear and the semiring is idempotent, commutative, and continuous. This result is applied to prove that the range of the otts transformation computed by a linear recognizable tree series transducer is pointwise recognizable.
The power of tree series transducers of type I and II
 PROC. 9TH INT. CONF. DEVELOPMENTS IN LANGUAGE THEORY, VOLUME 3572 OF LNCS
, 2005
"... The power of tree series transducers of type I and II is studied for IO as well as OI tree series substitution. More precisely, it is shown that the IO tree series transformations of type I (respectively, type II) are characterized by the composition of homomorphism topdown IO tree series transfor ..."
Abstract

Cited by 4 (4 self)
 Add to MetaCart
The power of tree series transducers of type I and II is studied for IO as well as OI tree series substitution. More precisely, it is shown that the IO tree series transformations of type I (respectively, type II) are characterized by the composition of homomorphism topdown IO tree series transformations with bottomup (respectively, linear bottomup) IO tree series transformations. On the other hand, polynomial OI tree series transducers of type I and II and topdown OI tree series transducers are equally powerful.