40351
|
Kohavi R. Bottom-up induction of oblivious read-once decision graphs. MACHINE LEARNING: ECML-94 1994. [DOI: 10.1007/3-540-57868-4_56] [Citation(s) in RCA: 14] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/12/2022]
|
40352
|
Maniezzo V. Genetic evolution of the topology and weight distribution of neural networks. ACTA ACUST UNITED AC 1994; 5:39-53. [DOI: 10.1109/72.265959] [Citation(s) in RCA: 316] [Impact Index Per Article: 10.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/09/2022]
|
40353
|
|
40354
|
Montague PR, Sejnowski TJ. The predictive brain: temporal coincidence and temporal order in synaptic learning mechanisms. Learn Mem 1994. [DOI: 10.1101/lm.1.1.1] [Citation(s) in RCA: 37] [Impact Index Per Article: 1.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
Abstract
Some forms of synaptic plasticity depend on the temporal coincidence of presynaptic activity and postsynaptic response. This requirement is consistent with the Hebbian, or correlational, type of learning rule used in many neural network models. Recent evidence suggests that synaptic plasticity may depend in part on the production of a membrane permeant-diffusible signal so that spatial volume may also be involved in correlational learning rules. This latter form of synaptic change has been called volume learning. In both Hebbian and volume learning rules, interaction among synaptic inputs depends on the degree of coincidence of the inputs and is otherwise insensitive to their exact temporal order. Conditioning experiments and psychophysical studies have shown, however, that most animals are highly sensitive to the temporal order of the sensory inputs. Although these experiments assay the behavior of the entire animal or perceptual system, they raise the possibility that nervous systems may be sensitive to temporally ordered events at many spatial and temporal scales. We suggest here the existence of a new class of learning rule, called a predictive Hebbian learning rule, that is sensitive to the temporal ordering of synaptic inputs. We show how this predictive learning rule could act at single synaptic connections and through diffuse neuromodulatory systems.
Collapse
|
40355
|
|
40356
|
What makes a problem hard for a genetic algorithm? Some anomalous results and their explanation. Mach Learn 1994. [DOI: 10.1007/bf00993046] [Citation(s) in RCA: 18] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/27/2022]
|
40357
|
|
40358
|
Farrell K, Mammone R, Assaleh K. Speaker recognition using neural networks and conventional classifiers. ACTA ACUST UNITED AC 1994. [DOI: 10.1109/89.260362] [Citation(s) in RCA: 150] [Impact Index Per Article: 5.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/10/2022]
|
40359
|
Bachmann CM, Musman SA, Luong D, Schultz A. Unsupervised BCM projection pursuit algorithms for classification of simulated radar presentations. Neural Netw 1994. [DOI: 10.1016/0893-6080(94)90047-7] [Citation(s) in RCA: 15] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022]
|
40360
|
Leng B, Buchanan BG, Nicholas HB. Protein secondary structure prediction using two-level case-based reasoning. J Comput Biol 1994; 1:25-38. [PMID: 8790451 DOI: 10.1089/cmb.1994.1.25] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 02/02/2023] Open
Abstract
We have developed a two-level case-based reasoning architecture for predicting protein secondary structure. The central idea is to break the problem into two levels: (i) reasoning at the object (protein) level and using the global information from this level to focus on a more restricted problem space; (ii) decomposing objects into pieces (segments) and reasoning at the level of internal structures. As a last step to the procedure, inferences from the parts of the internal structure are synthesized into predictions about global structure. The architecture has been developed and tested on a commonly used data set with 69.5% predictive accuracy. It was then tested on a new data set with 68.2% accuracy. With additional tuning, over 70% accuracy was achieved. In addition, a series of experiments were conducted to test various aspects of the method and the results are informative.
Collapse
Affiliation(s)
- B Leng
- Department of Computer Science, University of Pittsburgh, PA 15260, USA
| | | | | |
Collapse
|
40361
|
Fisher D, Ling Xu, Carnes J, Reich Y, Fenves J, Chen J, Shiavi R, Biswas G, Weinberg J. Applying AI clustering to engineering tasks. ACTA ACUST UNITED AC 1993. [DOI: 10.1109/64.248353] [Citation(s) in RCA: 13] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/09/2022]
|
40362
|
|
40363
|
Zeng Z, Goodman RM, Smyth P. Learning Finite State Machines With Self-Clustering Recurrent Networks. Neural Comput 1993. [DOI: 10.1162/neco.1993.5.6.976] [Citation(s) in RCA: 82] [Impact Index Per Article: 2.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/04/2022]
Abstract
Recent work has shown that recurrent neural networks have the ability to learn finite state automata from examples. In particular, networks using second-order units have been successful at this task. In studying the performance and learning behavior of such networks we have found that the second-order network model attempts to form clusters in activation space as its internal representation of states. However, these learned states become unstable as longer and longer test input strings are presented to the network. In essence, the network “forgets” where the individual states are in activation space. In this paper we propose a new method to force such a network to learn stable states by introducing discretization into the network and using a pseudo-gradient learning rule to perform training. The essence of the learning rule is that in doing gradient descent, it makes use of the gradient of a sigmoid function as a heuristic hint in place of that of the hard-limiting function, while still using the discretized value in the feedback update path. The new structure uses isolated points in activation space instead of vague clusters as its internal representation of states. It is shown to have similar capabilities in learning finite state automata as the original network, but without the instability problem. The proposed pseudo-gradient learning rule may also be used as a basis for training other types of networks that have hard-limiting threshold activation functions.
Collapse
Affiliation(s)
- Zheng Zeng
- Department of Electrical Engineering, 116-81, California Institute of Technology, Pasadena, CA 91125 USA
| | - Rodney M. Goodman
- Department of Electrical Engineering, 116-81, California Institute of Technology, Pasadena, CA 91125 USA
| | - Padhraic Smyth
- Jet Propulsion Laboratory, 238-420, California Institute of Technology, Pasadena, CA 91109 USA
| |
Collapse
|
40364
|
Mahadevan S, Mitchell TM, Mostow J, Steinberg L, Tadepalli PV. An apprentice-based approach to knowledge acquisition. ARTIF INTELL 1993. [DOI: 10.1016/0004-3702(93)90059-k] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/24/2022]
|
40365
|
|
40366
|
Schultz A, Grefenstette J, De Jong K. Test and evaluation by genetic algorithms. ACTA ACUST UNITED AC 1993. [DOI: 10.1109/64.236476] [Citation(s) in RCA: 25] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
|
40367
|
|
40368
|
Plausible inference and implicit representation. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00030934] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40369
|
Synchronization and cognitive carpentry: From systematic structuring to simple reasoning. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031095] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40370
|
|
40371
|
Rule acquisition and variable binding: Two sides of the same coin. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031058] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
|
40372
|
|
40373
|
What we know and the LTKB. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031113] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40374
|
Phase logic is biologically relevant logic. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031174] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40375
|
Imam IF, Michalski RS. Learning decision trees from decision rules: A method and initial results from a comparative study. J Intell Inf Syst 1993. [DOI: 10.1007/bf00962072] [Citation(s) in RCA: 6] [Impact Index Per Article: 0.2] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/30/2022]
|
40376
|
Psychological implications of the synchronicity hypothesis. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031137] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40377
|
Useful ideas for exploiting time to engineer representations. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031150] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40378
|
Ethereal oscillations. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031216] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40379
|
Not all reflexive reasoning is deductive. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x0003106x] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
|
40380
|
Could static binding suffice? Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00030946] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40381
|
Reasoning, learning and neuropsychological plausibility. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00030971] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
|
40382
|
Toward a unified behavioral and brain science. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031009] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
|
40383
|
Connectionism and syntactic binding of concepts. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00030983] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
|
40384
|
Self-organizing neural models of categorization, inference and synchrony. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031034] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/06/2022]
|
40385
|
Reflections on reflexive reasoning. Behav Brain Sci 1993. [DOI: 10.1017/s0140525x00031101] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/07/2022]
|
40386
|
Abstract
A genetic algorithm is a form of evolution that occurs on a computer. Genetic algorithms are a search method that can be used for both solving problems and modeling evolutionary systems. With various mapping techniques and an appropriate measure of fitness, a genetic algorithm can be tailored to evolve a solution for many types of problems, including optimization of a function of determination of the proper order of a sequence. Mathematical analysis has begun to explain how genetic algorithms work and how best to use them. Recently, genetic algorithms have been used to model several natural evolutionary systems, including immune systems.
Collapse
Affiliation(s)
- S Forrest
- Department of Computer Science, University of New Mexico, Albuquerque 87131-1386
| |
Collapse
|
40387
|
Roa R, Lu SY. A knowledge-based equation discovery system for engineering domains. ACTA ACUST UNITED AC 1993. [DOI: 10.1109/64.223989] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/05/2022]
|
40388
|
|
40389
|
Dayan P. Improving Generalization for Temporal Difference Learning: The Successor Representation. Neural Comput 1993. [DOI: 10.1162/neco.1993.5.4.613] [Citation(s) in RCA: 267] [Impact Index Per Article: 8.6] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/04/2022]
Abstract
Estimation of returns over time, the focus of temporal difference (TD) algorithms, imposes particular constraints on good function approximators or representations. Appropriate generalization between states is determined by how similar their successors are, and representations should follow suit. This paper shows how TD machinery can be used to learn such representations, and illustrates, using a navigation task, the appropriately distributed nature of the result.
Collapse
Affiliation(s)
- Peter Dayan
- Computational Neurobiology Laboratory, The Salk Institute, P.O. Box 85800, San Diego, CA 92186-5800 USA
| |
Collapse
|
40390
|
|
40391
|
Abstract
It is a striking fact that in humans the greatest learning occurs precisely at that point in time--childhood--when the most dramatic maturational changes also occur. This report describes possible synergistic interactions between maturational change and the ability to learn a complex domain (language), as investigated in connectionist networks. The networks are trained to process complex sentences involving relative clauses, number agreement, and several types of verb argument structure. Training fails in the case of networks which are fully formed and 'adultlike' in their capacity. Training succeeds only when networks begin with limited working memory and gradually 'mature' to the adult state. This result suggests that rather than being a limitation, developmental restrictions on resources may constitute a necessary prerequisite for mastering certain complex domains. Specifically, successful learning may depend on starting small.
Collapse
Affiliation(s)
- J L Elman
- Department of Cognitive Science, University of California, San Diego, La Jolla 92093-0526
| |
Collapse
|
40392
|
Bailin S, Henderson S, Truszkowski W. Application of machine learning to the organization of institutional software repositories. TELEMATICS AND INFORMATICS 1993. [DOI: 10.1016/0736-5853(93)90031-x] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/01/2022]
|
40393
|
Elkan C, Greiner R. Building large knowledge-based systems: Representation and inference in the cyc project. ARTIF INTELL 1993. [DOI: 10.1016/0004-3702(93)90092-p] [Citation(s) in RCA: 17] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/25/2022]
|
40394
|
|
40395
|
|
40396
|
Abstract
Computational models are useful tools for exploring the nature of human cognitive processes. In particular, connectionist models are providing researchers with new ways of thinking about the basic nature of cognition and its implementation in the brain. They support novel explanations of important aspects of perception, memory, language, thought and cognitive development, and allow cognitive processes to be linked with the underlying physiological mechanisms. The models also aid our understanding of how disorders of brain function lead to disorders of cognition.
Collapse
Affiliation(s)
- J L McClelland
- Department of Psychology, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213-3890
| | | |
Collapse
|
40397
|
Fisher D, Hapanyengwi G. Database management and analysis tools of machine induction. J Intell Inf Syst 1993. [DOI: 10.1007/bf01066545] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
|
40398
|
|
40399
|
A reply to Cohen's book review of Creating a Memory of Causal Relationships. Mach Learn 1993. [DOI: 10.1007/bf00993506] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
|
40400
|
BEN-DAVID ARIE. THE RELATION BETWEEN ORDINAL PROBLEM SPACE SIZES AND THE MAXIMUM NUMBER OF ORDINAL CLASSIFICATION RULES. Comput Intell 1993. [DOI: 10.1111/j.1467-8640.1993.tb00299.x] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
|