Arrow Research search

Author name cluster

Mark Derthick

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

5 papers
1 author row

Possible papers

5

AAAI Conference 1991 Conference Paper

A Minimal Encoding Approach to Feature Discovery

  • Mark Derthick

This paper discusses unsupervised learning of orthogonal concepts on relational data. Relational predicates, while formally equivalent to the features of the conceptlearning literature, are not a good basis for defining concepts. Hence the current task demands a much larger search space than traditional concept learning algorithms, the sort of space explored by connectionist algorithms. However the intended application, using the discovered concepts in the Cyc knowledge base, requires that the concepts be interpretable by a human, an ability not yet realized with connectionist algorithms. Interpretability is aided by including a characterization of simplicity in the evaluation function. For Hinton’ s Family Relations data, we do find cleaner, more intuitive features. Yet when the solutions are not known in advance, the difficulty of interpreting even features meeting the simplicity criteria calls into question the usefulness of any reformulation algorithm that creates radically new primitives in a knowledge-based setting. At the very least, much more sophisticated explanation tools are needed.

PDF Details

AIJ Journal 1990 Journal Article

Mundane reasoning by settling on a plausible model

  • Mark Derthick

Connectionist networks are well suited to everyday common sense reasoning. Their ability to simultaneously satisfy multiple soft constraints allows them to select from conflicting information in finding a plausible interpretation of a situation. However these networks are poor at reasoning using the standard semantics of classical logic, based on truth in all possible models. This article shows that using an alternate semantics, based on truth in a single most plausible model, there is an elegant mapping from theories expressed using the syntax of propositional logic onto connectionist networks. An extension of this mapping to allow for limited use of quantifiers suffices to build a network from knowledge bases expressed in a frame language similar to KL-ONE. Although finding optimal models of these theories is intractable, the networks admit a hill climbing search algorithm that can be tuned to give satisfactory answers in familiar situations. The article concludes with an example of retrieval involving incomplete and inconsistent information. Although this example works well, much remains before realistic domains are feasible.

Details DOI

AAAI Conference 1987 Conference Paper

Counterfactual Reasoning with Direct Models

  • Mark Derthick

Most of the effort AI has put into common sense reasoning has involved inference by sequential rule application. This approach is most effective in well characterized domains where any valid chain of inference from a set of observations leads to an acceptable interpretation. In more realistic cases where there are multiple consistent interpretations that are not equally good, or where there are no consistent interpretations, it seems more natural to choose the best alternative based on the interpretations themselves rather than the chains of inference used to derive them. ,uKLONE is a connectionist network which uses simulated annealing to search the space of interpretations, or models. Inconsistent theories lead to generation of models which come as close as possible to satisfying all of the axioms, so counterfactual reasoning can be accomplished by the same mechanism as factual reasoning. An example involving conflicting information is presented for which uKLONE finds an intuitively plausible interpretation.

PDF Details