Browse

ftp://nervous.cis.ohio-state.edu (49)

	ftp://nervous.cis.ohio-state.edu/pub/papers/91-JK-MAM.ps.Z, 19910319 1 Multiassociative Memory John F. Kolen Jordan B. Pollack The Laboratory for AI Research Department of Computer and Information Science The Ohio State University Columbus, OH, 43210 Telephone:(614)292-7402 Email: kolen-j@cis.ohio-state.edu
	ftp://nervous.cis.ohio-state.edu/pub/papers/89-JP-AIREVIEW.ps.Z, 19910725 Connectionism: Past, Present, and Future Jordan B. Pollack Computer & Information Science Department The Ohio State University 1. Introduction Research efforts to study computation and cognitive modeling on neurally-inspired mechanisms have come to be called Connectionism. Rather than being brand-new,
	ftp://nervous.cis.ohio-state.edu/pub/papers/89-JP-RECURSIVE.ps.Z, 19910726 Recursive Distributed Representations Jordan B. Pollack Laboratory for AI Research & Computer & Information Science Department The Ohio State University 2036 Neil Avenue Columbus, OH 43210 (614) 292-4890 pollack@cis.ohio-state.edu
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-pa-evinduct.ps.Z, 19920506
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-GS-PREEMPTRONS.ps.Z, 19920513
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-TB-PLANNING-COMPLEXITY.ps.Z, 19920707 e Comput tion l Complexity of Proposition l S RIPS Pl nnin Tom lan er Laboratory for Artificial Intelligence Research Department of Computer and Information Science The Ohio State University Columbus, Ohio USA email: byland cis.ohio-state.edu phone: 614-292-5835 May 15, 1992 submitted to Artificial
	ftp://nervous.cis.ohio-state.edu/pub/papers/gt-hist.ps.Z, 19920731 Generic Tasks: Evolution of An Idea B. Chandrasekaran Laboratory for AI Research (LAIR) The Ohio State University Columbus, OH 43210 DRAFT, JULY 30, 1992 Introduction For about a decade my coworkers and I at LAIR have been developing an approach to the analysis and construction of knowledge-based
	ftp://nervous.cis.ohio-state.edu/pub/papers/91-RF-PEYER-PEIRCE-REPORT.ps.Z, 19920812 Peirce-I : omain-Independent Problem Solver for bductive ssembl Richard Fox and John Josephson ay 19, 1992
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-RF-EVOLUTION.ps.Z, 19920812 c , , n o uc ion Abduction is inference to the best explanation. This reasoning method is very useful in accomplishing a variety of tasks such as diagnosis, identification, and recognition. However, it has also been used for solving theory formation and theory
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-RF-ARTREC.ps.Z, 19920812 n uctive rticulator eco nition ste ec nical e ort ichard ox, John Josephson and evin enzo pril , 1 Introduction We describe here an automated word recognition system that uses articulatory motions produced during human speech as input. The system uses a Layered Abductive problem solving strategy to
	ftp://nervous.cis.ohio-state.edu/pub/papers/91-RF-THREE-LEVEL.ps.Z, 19920812 ree evel c io ac i e or or eco i io ro r ic la io ichard ox, John Josephson and unil hadani a 1 , 1 2 Introduction Abduction is inference to the best explanation . This is a form of reasoning used to account for the appearance of some set of findings, or to causally explain them. Abduction
	ftp://nervous.cis.ohio-state.edu/pub/papers/91-RF-CV.ps.Z, 19920812 x eri ent an esults ichard ox and John Josephson ugust 12, 1992 Introduction We have posited a method for attempting speech recognition by layered abduction. Abduction is a reasoning process of explaining findings by hypothesization. The task-level description of abduction matches that of speech
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-RF-DISSERTATION.ps.Z, 19920813 B. Chandrasekaran John Josephson Osamu Fujimura Terry Patten Advisor Co-advisor Department of Computer and Information Science c Copyright by Richard Keith Fox 1992 In loving memory of my mother ii o l g There are many people whom I am indebted to. I wish to thank my co-advisors, B. Chandrasekaran
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-DA-SISYPHUS.ps.Z, 19921014 Acquiring Knowledge of Knowledge Acquisition: A self-study of Generic Tasks1 Dean Allemang2 Thomas E. Rothenfluh3 92-DA-SISYPHUS 1Appeared also in: Wetter, T., Althoff, K.-D., Gaines, B.R., Linster, M., & Schmalhofer, F. (Eds.) (1992) Current Developments in Knowledge Acquisition - EKAW'92. Proceedings
	ftp://nervous.cis.ohio-state.edu/pub/papers/ijcai93.ps, 19921127 Learning With a Competetive Population Angeline & Pollack The Ohio State University November 13, 1992 9 Epstein, S. In search of the ideal trainer , to appear. Epstein, S. Learning expertise from the opposition: the role of the trainer in a competetive environment. , The
	ftp://nervous.cis.ohio-state.edu/pub/papers/alife92.ps, 19921127 Coevolving High-level Representations Peter J. Angeline and Jordan B. Pollack LAIR Technical Report 92-PA-COEVOLVE Submitted to the Proceedings of Artificial Life III Laboratory for Artificial Intelligence Research Computer and Information Science Department The Ohio State University Columbus, Ohio
	ftp://nervous.cis.ohio-state.edu/pub/papers/visual-symposium-report.ps, 19921214 Reasoning with Diagrammatic Representations: A Report on the AAAI Spring Symposium, March 25-27, 1992 B. Chandrasekaran Laboratory for AI Research The Ohio State University Columbus, OH 43210 N. Hari Narayanan Hitachi Advanced Research Laboratory Hatoyama, Saitama 350-03, Japan and Yumi Iwasaki
	ftp://nervous.cis.ohio-state.edu/pub/papers/design-rationale.ps, 19930128 Functional Representation as Design Rationale B. Chandrasekaran Ashok Goel Yumi Iwasaki Laboratory for AI Research Department of Computer Science Knowledge Systems Lab 2036 Neil Ave Georgia Institute of Technology Stanford University The Ohio State University Atlanta, GA 30332 701 Welch Road Columbus,
	ftp://nervous.cis.ohio-state.edu/pub/papers/ai-quals-reading-list-stanford.ps, 19930222 Reading List for the Qualifying Examination in Artificial Intelligence Karen Myers Devika Subramanian Ramin Zabih 1989 This is the version of the reading list for the Stanford Qualifying examination in Artificial Intelligence of Spring, 1989. Please send comments to ronnyk@CS.Stanford.EDU Overview The
	ftp://nervous.cis.ohio-state.edu/pub/papers/93-pa-ep93.ps.Z, 19930224 Evolutionary Module Acquisition Angeline and Pollack The Ohio State University February 24, 1993 10 9. References Angeline, P. (1993) An analysis of evolutionary algorithms , Submitted to International Conference on Genetic Algorithms 1993. Angeline, P. and Pollack, J. (1993) Coevolving high-level
	ftp://nervous.cis.ohio-state.edu/pub/papers/cog-arch.ps, 19930325 Mss dated February 28, 1993 Architecture of Intelligence: The Problems and Current Approaches to Solutions1 B. Chandrasekaran and Susan G. Josephson Laboratory for AI Research The Ohio State University Columbus, OH 43210 USA
	ftp://nervous.cis.ohio-state.edu/pub/papers/92-pa-coevolve.ps.Z, 19930427 Coevolving High-Level Representations Peter J. Angeline and Jordan B. Pollack Laboratory for Artificial Intelligence Research Computer and Information Science Department The Ohio State University Columbus, Ohio 43210 pja@cis.ohio-state.edu pollack@cis.ohio-state.edu To Appear in: Artificial Life III The
	ftp://nervous.cis.ohio-state.edu/pub/papers/93-pa-compfit.ps.Z, 19930505 Competetive Environments Evolve Better Solutions for Complex Tasks Angeline and Pollack The Ohio State University May 5, 1993 7 constructed. The adaptability of the content of the single population may be more beneficial to the evolutionary development of solutions than a predetermined bipartite
	ftp://nervous.cis.ohio-state.edu/pub/papers/91-pa-avfusion.ps.Z, 19930604 voiding Fusion in Floating Symbol Systems Peter J. Angeline and Jordan B. Pollack Laboratory for Artificial Intelligence esearch Department of Computer and Information Sciences The Ohio State University Columbus, Ohio 43210 pja cis.ohio-state.edu pollack cis.ohio-state.edu June 4, 1993 1 ummar One of
	ftp://nervous.cis.ohio-state.edu/pub/papers/91-pa-hraams.ps.Z, 19930604 The Ohio State University June 4, 1993 1 Hierarchical RAAMs: A Uniform Modular Architecture Peter J. Angeline and Jordan B. Pollack Laboratory for Artificial Intelligence Research Department of Computer and Information Science The Ohio State University Columbus, Ohio 43210 pja@cis.ohio-state.edu
	ftp://nervous.cis.ohio-state.edu/pub/papers/93-pa-gnarly.ps.Z, 19930716 An Evolutionary Algorithm that Constructs Recurrent Neural Networks Peter J. Angeline, Gregory M. Saunders and Jordan B. Pollack Laboratory for Artificial Intelligence Research Computer and Information Science Department The Ohio State University Columbus, Ohio 43210 pja@cis.ohio-state.edu
	ftp://nervous.cis.ohio-state.edu/pub/papers/93-JK-PARADOX.ps.Z, 19930819 The Observers Paradox: Apparent Computational Complexity in Physical Systems John F. Kolen and Jordan B. Pollack To appear in The Journal of Experimental and Theoretical Artificial Intellignce Running Head: The Observers Paradox August 15, 1993 Laboratory for Artificial Intelligence Research Department
	ftp://nervous.cis.ohio-state.edu/pub/papers/fr-paper.ps, 19930826 08/26/93 The Functional Representation Language: A Framework for Reasoning About Functions and Causal Processes of Devices B. Chandrasekaran Laboratory for AI Research The Ohio State University Columbus, OH 43210 Table of Contents The Functional Representation Language:
	ftp://nervous.cis.ohio-state.edu/pub/papers/93-pa-emerge.ps, 19931007 Genetic Programming and Emergent Intelligence Angeline Genetic Programming and Emergent Intelligence Peter J. Angeline Laboratory for Artificial Intelligence Research (LAIR) Computer and Information Sciences Department The Ohio State University Columbus, OH 43210 pja@cis.ohio-state.edu To appear as a
	ftp://nervous.cis.ohio-state.edu/pub/papers/jk-foolsgold.ps.Z, 19931007 Fool s Gold: Extracting Finite State Machines From Recurrent Network Dynamics John F. Kolen Laboratory for AI Research Department of Computer and Information Science The Ohio State University Columbus, OH 43202 kolen-j@cis.ohio-state.edu DRAFT: To appear in NIPS6
	ftp://nervous.cis.ohio-state.edu/pub/papers/aai-j-fr-paper.ps, 19931027 Applied Artificial Intelligence, Special Issue on Functional Reasoning. (mss dated 1/27/93) Functional Representation: A Brief Historical Perspective1 B. Chandrasekaran Laboratory for AI Research The Ohio State University Columbus, OH 43210
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter8.ps.Z, 19931201 144 CHAPTER VIII SUMMARY AND CONCLUSIONS 8.0 Summary and Conclusions The methodology of knowledge-based AI relies on a central assumption concerning the nature of intelligence: The paradigm has traditionally assumed that the symbolic, algorithmic character of the macrophenomena
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter5.ps.Z, 19931201 96 CHAPTER V EMERGENCE OF TASK-DIRECTED COMPONENT MANIPULATION 5.0 Emergence of Task-Directed Component Manipulation In the previous chapter, the natural dynamics of an evolutionary weak method were used to induce both the architecture and parametric values for a neural network. The demonstrated
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter3.ps.Z, 19931201 41 CHAPTER III THE EVOLUTIONARY WEAK METHOD AND EMERGENT INTELLIGENCE 3.0 The Evolutionary Weak Method and Emergent Intelligence The commonalities of evolutionary algorithms combined with their lack of task-specific knowledge indicates their status as a weak method which this dissertation calls the
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/dissrefs.ps.Z, 19931201 152 BIBLIOGRAPHY Ackley, D. H., G. E. Hinton and T. J. Sejnowski (1985). A learning algorithm for Boltzmann machines. Cognitive Science, 9, pp. 147-169. Agre, P.E. and D. Chapman (1987). Pengi: An implementation of a theory of activity. In Proceedings of the Sixth National Conference on Artificial
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter4.ps.Z, 19931201 65 CHAPTER IV THE EMERGENCE OF TASK-SPECIFIC STRUCTURES 4.0 The Emergence of Task-Specific Structures The induction of appropriate connectionist network architectures is a topic of current research. Most of the work by connectionists favor assuming an architecture for the network and simply learning the
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter1.ps.Z, 19931201 1 CHAPTER I INTRODUCTION: SEARCH AND EXPLICIT KNOWLEDGE 1.0 Introduction: Search and Explicit Knowledge Two broad goals cover most of the subtopics within artificial intelligence (AI) research (Schank 1987). One goal is the study of human intelligence and problem solving by creating computational models
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter0.ps.Z, 19931201 EVOLUTIONARY ALGORITHMS AND EMERGENT INTELLIGENCE DISSERTATION Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University By Peter John Angeline, B.S., M.S. The Ohio State University 1993 Dissertation Committee: Approved
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter6.ps.Z, 19931201 104 CHAPTER VI THE EMERGENCE OF PROBLEM DECOMPOSITIONS AND HIGH-LEVEL REPRESENTATIONS 6.0 The Emergence of Problem Decompositions and High-Level Representations In this chapter, modular programs are induced using genetic program with additional mutation operators designed to allow decompositions to
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter2.ps.Z, 19931201 15 CHAPTER II BACKGROUND 2.0 Background This chapter describes background relating to weak methods, strong methods and evolutionary algorithms. It is argued that evolutionary algorithms consist a new type of weak method, called the evolutionary weak method, that adapts its task-independent method of
	ftp://nervous.cis.ohio-state.edu/pub/papers/DISS/pja/chapter7.ps.Z, 19931204 125 CHAPTER VII EMERGENT GOAL-DIRECTED BEHAVIOR 7.0 Emergent Goal-Directed Behavior The task environment is an important component of emergent intelligent systems. Often in computational problem solving, the feedback to a method comes from an objective function. Typically, these objective functions
	ftp://nervous.cis.ohio-state.edu/pub/papers/93-GS-STREVOL.ps.Z, 19931209 Gregory M. Saunders, Peter J. Angeline, and Jordan B. Pollack Laboratory for Artificial Intelligence Research Department of Computer and Information Science The Ohio State University Columbus, Ohio 43210 saunders@cis.ohio-state.edu
	ftp://nervous.cis.ohio-state.edu/pub/papers/kolen.dis.ps.Z, 19940126 Very few beings really seek knowledge in this world. Few really ask. On the contrary, they try to wring from the unknown the answers they have already shaped in their own minds justification, confirmation, forms of consolation without which they can t go on. To really ask is to open the door to a
	ftp://nervous.cis.ohio-state.edu/pub/papers/p62.ps, 19940805 62 CHAPTER IV Figure 20: Several fractal sets generated by iterated function systems. Sierpinski Triangle Black Spleenwort Fern Fractal SpiralsFractal T ree
	ftp://nervous.cis.ohio-state.edu/pub/papers/kolen.thesis.ps.Z, 19940812
	ftp://nervous.cis.ohio-state.edu/pub/papers/94-GS-EVCOMM.ps.gz, 19940824 The Evolution of Communication in Adaptive Agents Gregory M. Saunders and Jordan B. Pollack Laboratory for Artificial Intelligence Research Department of Computer and Information Science The Ohio State University Columbus, Ohio 43210 USA saunders@cis.ohio-state.edu pollack@cis.ohio-state.edu Phone:
	ftp://nervous.cis.ohio-state.edu/pub/papers/control-at-knowledge-level.ps.Z, 19940824 Understanding Control at the Knowledge Level B. Chandrasekaran Laboratory for AI Research The Ohio State University Columbus, OH 43210 Email: chandra@cis.ohio-state.edu
	ftp://nervous.cis.ohio-state.edu/pub/papers/94-GS-SAB.ps.gz, 19940831
	ftp://nervous.cis.ohio-state.edu/pub/papers/kolen.a4.ps, 19941103 October 31, 1994 2 Proinde cum venabere, licebit, auctore me, ut panarium et lagunculam sic etiam pugillares feras. Mirum est ut animus agitatione motuque corporis excitetut. Ad retia sedebam: erat in proximo non venabulum aut lancea, sed stilus et pugilares; meditabar aliquid enotabamque, plenas tamen