http:^^www.cs.utexas.edu^users^ml^abstracts.html

This data set contains WWW-pages collected from computer science departments of various universities

<blockquote> 

This paper presents results on using a new inductive logic programming
method called FOIDL to learn the past tense of English verbs. The past
tense task has been widely studied in the context of the
symbolic/connectionist debate.  Previous papers have presented results
using various neural-network and decision-tree learning methods.  We
have developed a technique for learning a special type of Prolog
program called a <em>first-order decision list</em>, defined as an
ordered list of clauses each ending in a cut.  FOIDL is based on FOIL
(Quinlan, 1990) but employs intensional background knowledge and
avoids the need for explicit negative examples.  It is particularly
useful for problems that involve rules with specific exceptions, such
as the past-tense task.  We present results showing that FOIDL learns
a more accurate past-tense generator from significantly fewer examples
than all other previous methods.
</blockquote>

<!WA32><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/foidl-bkchapter-95.ps.Z">
<!WA33><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>

<! ===========================================================================>

<a name="dolphin-ewsp-95.ps.Z" </a>

<b> <li> Hybrid Learning of Search Control for Partial-Order Planning <br> </b>

Tara A. Estlin and Raymond J. Mooney <br>

<cite>New Directions in AI Planning</cite>, M. Ghallab and A. Milani, Eds,
IOS Press, 1996, pp. 129-140. <p>  

<blockquote>

This paper presents results on applying a version of the DOLPHIN
search-control learning system to speed up a partial-order planner.
DOLPHIN integrates explanation-based and inductive learning techniques
to acquire effective clause-selection rules for Prolog programs.  A
version of the UCPOP partial-order planning algorithm has been
implemented as a Prolog program and DOLPHIN used to automatically
learn domain-specific search control rules that help eliminate
backtracking. The resulting system is shown to produce significant
speedup in several planning domains.

</blockquote>

<!WA34><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/dolphin-ewsp-95.ps.Z">
<!WA35><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>

<! ===========================================================================>

<a name="banner-icnn-96.ps.Z"</a>


<b><li>Revising Bayesian Network Parameters Using Connectionist Methods<br></b>
Sowmya Ramachandran and Raymond J. Mooney<br>

<cite>Proceedings of the International Conference on Neural
Networks (ICNN-96)</cite>, Special Session on Knowledge-Based Artificial
Neural Networks, Washington DC, June 1996. <p>

<blockquote>
The problem of learning Bayesian networks with hidden variables is known to
be a hard problem. Even the simpler task of learning just the conditional
probabilities on a Bayesian network with hidden variables is hard. In this
paper, we present an approach that learns the conditional probabilities on
a Bayesian network with hidden variables by transforming it into a
multi-layer feedforward neural network (ANN). The conditional probabilities
are mapped onto weights in the ANN, which are then learned using standard
backpropagation techniques. To avoid the problem of exponentially large
ANNs, we focus on Bayesian networks with noisy-or and noisy-and
nodes. Experiments on real world classification problems demonstrate the
effectiveness of our technique.
</blockquote>

<!WA36><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/banner-icnn-96.ps.Z">
<!WA37><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>


<! ===========================================================================>

<a name="qdocs-dissertation-95.ps.Z"</a>

<b><li>Qualitative Multiple-Fault Diagnosis of Continuous Dynamic Systems Using
Behavioral Modes<br></b>

Siddarth Subramanian<br>

Ph.D. Thesis, Department of Computer Sciences, University of Texas at Austin, August, 1995.<p>

<blockquote>

As systems like chemical plants, power plants, and automobiles get
more complex, online diagnostic systems are becomingly increasingly
important.  One of the ways to rein in the complexity of describing
and reasoning about large systems such as these is to describe them
using qualitative rather than quantitative models.<p>

Model-based diagnosis is a class of diagnostic techniques that use
direct knowledge about how a system functions instead of expert rules
detailing causes for every possible set of symptons of a broken
system.  Our research builds on standard methods for model-based
diagnosis and extends them to the domain of complex dynamic systems
described using qualitative models.<p>

We motivate and describe out algorithm for diagnosing faults in a
dynamic system given a qualitative model and a sequence of qualitative
states.  The main contributions in this algorithm include a method for
propagating dependencies while solving a general constraint
satisfaction problem, and a method for verfying the compatibility of a
behavior with a model across time.  The algorithm can diagnose
multiple faults and uses models of faulty behavior, or behavioral
modes.<p>

We then demonstrate these techniques using an implemented program
called QDOCS and test it on some realistic problems.  Through our
experiments with a model of the reaction control system (RCS) of the
space shuttle and with a level-controller for a reaction tank, we show
that QDOCS demonstrates the best balance of generality, accuracy and
efficiency among known systems.<p>
</blockquote>

<!WA38><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/qdocs-dissertation-95.ps.Z">
<!WA39><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>

<! ===========================================================================>

<a name="chill-dissertation-95.ps.Z"</a>

<b><li>Using Inductive Logic Programming to Automate the Construction of 
Natural Language Parsers<br></b>

John M. Zelle<br>

Ph.D. Thesis, Department of Computer Sciences, University of Texas at Austin, August, 1995.<p>

<blockquote>
Designing computer systems to understand natural language input is a
difficult task.  In recent years there has been considerable interest
in corpus-based methods for constructing natural language parsers.
These empirical approaches replace hand-crafted grammars with
linguistic models acquired through automated training over language
corpora.  A common thread among such methods to date is the use of
propositional or probablistic representations for the learned
knowledge.  This dissertation presents an alternative approach based
on techniques from a subfield of machine learning known as inductive
logic programming (ILP).  ILP, which investigates the learning of
relational (first-order) rules, provides an empirical method for
acquiring knowledge within traditional, symbolic parsing frameworks.<p>

This dissertation details the architecture, implementation and
evaluation of CHILL a computer system for acquiring natural
language parsers by training over corpora of parsed text.  CHILL
treats language acquisition as the learning of search-control rules
within a logic program that implements a shift-reduce parser.  Control
rules are induced using a novel ILP algorithm which handles difficult
issues arising in the induction of search-control heuristics.  Both
the control-rule framework and the induction algorithm are crucial to
CHILL's success.<p>

The main advantage of CHILL over propositional counterparts is
its flexibility in handling varied representations.  CHILL has
produced parsers for various analyses including case-role mapping,
detailed syntactic parse trees, and a logical form suitable for
expressing first-order database queries.  All of these tasks are
accomplished within the same framework, using a single, general
learning method that can acquire new syntactic and semantic categories
for resolving ambiguities.<p>

Experimental evidence from both aritificial and real-world corpora
demonstrate that CHILL learns parsers as well or better than
previous artificial neural network or probablistic approaches on
comparable tasks.  In the database query domain, which goes beyond the
scope of previous empirical approaches, the learned parser outperforms
an existing hand-crafted system.  These results support the claim that
ILP techniques as implemented in CHILL represent a viable
alternative with significant potential advantages over neural-network,
propositional, and probablistic approaches to empirical parser
construction.<p>
</blockquote>

<!WA40><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/chill-dissertation-95.ps.Z">
<!WA41><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>


<! ===========================================================================>

<a name="chill-compling-95.ps.Z"</a>

<b><li>An Inductive Logic Programming Method for Corpus-based Parser
Construction<br></b>

John M. Zelle and Raymond J. Mooney<br>

Submitted to Computational Linguistics<p>

<blockquote>
In recent years there has been considerable research into corpus-based
methods for parser construction.  A common thread in this research has
been the use of propositional representations for learned knowledge.
This paper presents an alternative approach based on techniques from a
subfield of machine learning known as inductive logic programming
(ILP).  ILP, which investigates the learning of relational
(first-order) rules, provides a way of using empricial methods to
acquire knowledge within traditional, symbolic parsing frameworks.  We
describe a novel method for constructing deterministic Prolog parsers
from corpora of parsed sentences.  We also discuss several advantages
of this approach compared to propositional alternatives and present
experimental results on learning complete parsers using several
corpora including the ATIS corpus from the Penn Treebank.
</blockquote>

<!WA42><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/chill-compling-95.ps.Z">
<!WA43><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>

<! ===========================================================================>

<a name="chill-ijcai-nll-95.ps.Z"</a>

<b><li>A Comparison of Two Methods Employing Inductive Logic Programming for
Corpus-based Parser Constuction<br></b>

John M. Zelle and Raymond J. Mooney<br>

<cite>Working Notes of the IJCAI-95 Workshop on New Approaches to
Learning for Natural Language Processing</cite>, pp.79-86, Montreal,
Quebec, August, 1995.<p>

<blockquote> 
This paper presents results from recent experiments with CHILL,
a corpus-based parser acquisition system.  CHILL treats grammar
acquisition as the learning of search-control rules within a logic
program.  Unlike many current corpus-based approaches that use
propositional or probabilistic learning algorithms, CHILL uses
techniques from inductive logic programming (ILP) to learn relational
representations.  The reported experiments compare CHILL's
performance to that of a more naive application of ILP to parser
acquisition.  The results show that ILP techniques, as employed in
CHILL, are a viable alternative to propositional methods and
that the control-rule framework is fundamental to CHILL's
success.
</blockquote>

<!WA44><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/chill-ijcai-nll-95.ps.Z">
<!WA45><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>



<! ===========================================================================>

<a name="chill-ifoil-ml-95.ps.Z"</a>

<b><li>Inducing Logic Programs without Explicit Negative Examples<br></b>

John M. Zelle, Cynthia A. Thompson, Mary Elaine Califf, and Raymond J. Mooney<br>

<cite>Proceedings of the Fifth International Workshop on Inductive Logic
Programming</cite>, Leuven, Belguim, Sepetember 1995.<p>


<blockquote> This paper presents a method for learning logic programs
without explicit negative examples by exploiting an assumption of
<i>output completeness</i>. A mode declaration is supplied for the
target predicate and each training input is assumed to be accompanied
by all of its legal outputs.  Any other outputs generated by an
incomplete program implicitly represent negative examples; however,
large numbers of ground negative examples never need to be generated.
This method has been incorporated into two ILP systems, CHILLIN and
IFOIL, both of which use intensional background knowledge.  Tests on
two natural language acquisition tasks, case-role mapping and
past-tense learning, illustrate the advantages of the approach.
</blockquote>

<!WA46><a href="file://ftp.cs.utexas.edu/pub/mooney/papers/chill-ifoil-ml-95.ps.Z">
<!WA47><img align=top src="http://www.cs.utexas.edu/users/ml/paper.xbm"></a><p>

<! ===========================================================================>

<a name="wolfie-ack-95.ps.Z"</a>

<b><li>Acquisition of a Lexicon from Semantic Representations of Sentences<br></b>