Browsing by Author "Reggia, James"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item An External Tabletop Environment for an Interactive Brain Model(2007-09) Monner, Derek; Reggia, JamesAs an important step towards creating a biologically-inspired model of language learning in the human brain, we have created an external environment in which to embody such a model. It is our hope that the learning in such a model will be enhanced and accelerated by the grounding effects that such an environment can provide. The environment can be easily configured to approximate common tabletop testing environments used in clinical studies of human aphasia patients, which will allow us to draw parallels between said studies and similar computational experiments to be conducted by artificially lesioning our language-learning brain model. This document describes the capabilities and programming interfaces of the external environment and how a computational agent might interact with it.Item Identifying Fixed Points in Recurrent Neural Networks using Directional Fibers: Supplemental Material on Theoretical Results and Practical Aspects of Numerical Traversal(2016-12-12) Katz, Garrett; Reggia, JamesFixed points of recurrent neural networks can represent many things, including stored memories, solutions to optimization problems, and waypoints along non-fixed attractors. As such, they are relevant to a number of neurocomputational phenomena, ranging from low-level motor control and tool use to high-level problem solving and decision making. Therefore, global solution of the fixed point equations can improve our understanding and engineering of recurrent neural networks. While local solvers and statistical characterizations abound, we do not know of any method for efficiently and precisely locating all fixed points of an arbitrary network. To solve this problem we have proposed a novel strategy for global fixed point location, based on numerical traversal of mathematical objects we defined called directional fibers [2]. This report supplements our results in [2] by presenting certain technical aspects of our method in more depth.Item Integrating Knowledge-Based and Case-Based Reasoning(2006-08-30) Chabuk, Timur; Seifter, Mark; Salasin, John; Reggia, JamesThere has been substantial recent interest in integrating knowledge based reasoning (KBR) and case-based reasoning (CBR) within a single system due to the potential synergisms that could result. Here we describe our recent work investigating the feasibility of a combined KBR-CBR application-independent system for interpreting multi-episode stories/narratives, illustrating it with an application in the domain of interpreting urban warfare stories. A genetic algorithm is used to derive weights for selection of the most relevant past cases. In this setting, we examine the relative value of using input features of a problem for case selection versus using features inferred via KBR, versus both. We find that using both types of features is best (compared to human selection), but that input features are most helpful and inferred features are of marginal value. This finding supports the idea that KBR and CBR provide complimentary rather than redundant information, and hence that their combination in a single system is likely to be useful.