NIPS*95 Program

SUN NOV 26
----------

18:00-22:00	Registration


MON NOV 27
----------


08:30-18:00	Registration

09:30-17:30	Tutorials

18:30		Reception and Conference Banquet

20:30		Origins and future of flight:  A paleoecological perspective
		(BANQUET TALK)
		John H. McMasters
		Boeing Commercial Aircraft Company


TUE NOV 28
----------


		Oral Session 1:  Cognitive Science

08:30		Brain organization for language in children and adults
		(INVITED TALK)
		Elizabeth Bates
		UC San Diego

09:00		Learning the features of similarity (CS265)
		J. B. Tenenbaum
		Massachusetts Institute of Technology

09:20		A model of spatial representations in parietal cortex explains 
		hemineglect (CS328)
		A. Pouget, T. J. Sejnowski
		University of California, Los Angeles

09:40		The curse of dimensionality and human reading (CS192,
Spotlight)
		G. L. Martin
		MCC

09:45		Extracting tree-structured representations of trained
		networks (CS351)
		M. W. Craven, J. W. Shavlik
		University of Wisconsin, Madison


10:05-10:35	Break


		Oral Session 2:  Theory

10:35		Learning model bias (LT183)
		J. Baxter
		University of London

10:55		Statistical theory of overtraining -- Is cross-validation
		asymptotically effective? (LT268)
		S. Amari, N. Murata, K. Mueller, M. Finke, H. Yang
		GMD First

11:15		A bound on the error of cross validation using the
		approximation and estimation rates, with consequences for
		the training-test split (LT147, Spotlight)
		M. Kearns
		AT&T Bell Laboratories

		Learning with ensembles:  How overfitting can be 
		useful (LT262, Spotlight)
		P. Sollich, A. Krogh
		NORDITA

11:20		Neural networks with quadratic VC dimension (LT191)
		P. Koiran, E. D. Sontag
		LIP, ENS Lyon - CNRS

11:40		Learning recurrent perceptron mappings (LT210)
		B. Dasgupta, E. D. Sontag
		Rutgers University


12:00-14:00	Lunch


		Oral Session 3:  Neuroscience

14:00		Mapping brain function with functional magnetic resonance 
		imaging (INVITED TALK)
		Bruce Rosen
		Massachusetts General Hospital

14:30		Modeling interactions of the rat's place and head direction
		systems (NS140)
		A. D. Redish, D. S. Touretzky
		Carnegie Mellon University

14:50		Symmetry, inhibition, and correlation in spike trains of
		the motion area MT (NS162)
		W. Bair, C. Koch, E. Zohary
		California Institute of Technology

15:10		Information through a spiking neuron (NS294)
		C. Stevens, A. Zador
		The Salk Institute

15:30		Reorganization of somatosensory cortex after tactile
		training (NS333, Spotlight)
		R. S. Peterson, J. G. Taylor
		King's College London

		A dynamical model of context dependencies for the 
		vestibulo-ocular reflex (NS452, Spotlight)
		O. J. M. D. Coenen, T. J. Sejnowski
		The Salk Institute


15:35-16:05	Break


		Oral Session 4:  Speech and Signal Processing

16:05		Onset-based sound segmentation (SP103, Spotlight)
		L. S. Smith
		University of Stirling

		Laterally interconnected self-organizing maps in handwritten 
		digit rcognition (SP391, Spotlight)
		Y. Choe, J. Sirosh, R. Miikkulainen
		University of Texas, Austin

16:10		Forward-backward retraining of recurrent neural 
		networks (SP360)
		A. Senior, T. Robinson
		IBM

16:30		Context-Dependent classes in a hybrid recurrent network-HMM
		speech recognition system (SP407)
		D. J. Kershaw, M. M. Hochberg, A. J. Robinson
		Cambridge University


		Oral Session 5:  Algorithms and Architectures I


16:50		Adaptive mixture of probabilistic transducers (AA224)
		Y. Singer
		AT&T Bell Laboratories

17:10		REMAP: Recursive Estimation and Maximization of A
		posteriori Probabilities -- Application to transition-based
		connectionist speech recognition (AA94, Spotlight)
		Y. Konig, H. Bourlard, N. Morgan
		International Computer Science Institute

		Recurrent neural networks for missing or asynchronous data
		(AA238, Spotlight)
		Y. Bengio, F. Gingras
		Universite de Montreal

		Family discovery (AA311, Spotlight)
		S. M. Omohundro
		NEC Research Institute

		Discriminant adaptive nearest neighbor classification and
		regression (AA264, Spotlight)
		T. Hastie, R. Tibshirani
		Stanford University


WED NOV 29
----------


		Oral Session 6:  Algorithms & Architectures II

08:30		Learning Bayesian networks (INVITED TALK)
		David Heckerman
		Microsoft

09:00		Discovering structure in continuous variables using
		Bayesian networks (AA297)
		R. Hofmann, V. Tresp
		Siemens AG

09:20		Using pairs of data points to define splits for decision
		trees (AA396, Spotlight)
		G. E. Hinton, M. Revow
		University of Toronto


		Oral Session 7:  Implementations Spotlights

09:25		Silicon models for auditory scene analysis (IM378, Spotlight)
		J. Lazzaro, J. Wawrzynek
		University of California, Berkeley

		A visual smooth pursuit tracking chip (IM459, Spotlight)
		R. Etienne-Cummings, J. Van der Spiegel
		Southern Illinois University
 
		Model matching and SFMD computation (IM219, Spotlight)
		S. Rehfuss, D. Hammerstrom
		Oregon Graduate Institute 


09:35-10:05	Break


		Oral Session 8:  Vision

10:05		Classifying facial action (VS325)
		M. S. Bartlett, P. A. Viola, T. J. Sejnowski, J. Larsen, 
		J. C. Hager, P. Ekman
		The Salk Institute

10:25		Modeling saccadic targeting in visual search (VS367)
		R. P. N. Rao, G. J. Zelinsky, M. M. Hayhoe, D. H. Ballard
		University of Rochester

10:45		A model of transparent motion and nontransparent motion
		aftereffects (VS7)
		A. Grunewald
		Max Planck Institut fuer Biologische Kybernetik

11:05		A neural network model of 3D lightness perception (VS70,
		Spotlight)
		L. Pessoa, W. Ross
		Boston University

		Empirical entropy manipulation for real-world problems
		(VS176, Spotlight)
		P. Viola, N. N. Schraudolph, T. J. Sejnowski
		The Salk Institute 


		Oral Session 9:  Theory 

11:10		Optimization principles for the neural code (LT363, Spotlight)
		M. DeWeese, W. Bialek
		The Salk Institute

		Strong unimodality and efficient learning of \mu-Perceptron
		networks (LT375, Spotlight)
		M. Marchand, S. Hadjifaradji
		University of Ottawa

		Active learning in multilayer perceptrons (LT141, Spotlight)
		K. Fukumizu
		Ricoh Corp.

11:20		Dynamics of on-line gradient descent learning for
		multilayer neural networks (LT260)
		D. Saad, S. A. Solla
		The Niels Bohr Institute

11:40		Worst-case loss bounds for sigmoided neurons (LT381)
		D. P. Helmbold, J. Kivinen, M. K. Warmuth
		University of California, Santa Cruz


12:00-14:00	Lunch


13:00-14:00	Feedback session with NIPS board


		Oral Session 10:  Applications

14:00		Application of neural networks in the chemical process
		industries (INVITED TALK)
		Thomas McAvoy
		University of Maryland

14:30		A neural network autoassociator for induction motor failure
		prediction (AP288)
		T. Petsche, A. Marcantonio, C. Darken, S. J. Hanson, 
		G. M. Kuhn, I. Santoso
		Siemens Corporate Research

14:50		Using feedforward neural networks to monitor alertness from
		changes in EEG correlation and coherence (AP428)
		S. Makeig, T. Jung, T. J. Sejnowski
		Naval Health Research Center

15:10		A neural network classifier for the I1000 OCR chip (AP221,
		Spotlight)
		J. C. Platt, T. P. Allen
		Synaptics

		Predictive Q-routing:  A memory-based reinforcement learning
		approach to adaptive traffic control (AP129, Spotlight)
		S. P. M. Choi, D. Yeung
		The Hong Kong University of Science and Technology


15:15-15:40	Break


		Oral Session 11:  Control and Navigation

15:40		Improving elevator performance using reinforcement learning
		(CN195)
		R. H. Crites, A. G. Barto
		University of Massachusetts, Amherst

16:00		High-performance job-shop scheduling with a time-delay
		TD(lambda) network (CN387, Spotlight)
		W. Zhang, T. G. Dietterich
		Oregon State University

16:05		Competence acquisition in an autonomous mobile robot using
		hardware neural techniques (IM13)
		G. Jackson, A. F. Murray
		University of Edinburgh

16:25		Generalization in reinforcement learning:  Successful
		examples using sparse coarse coding (CN323)
		R. S. Sutton
		Stow Research

16:45		Stable linear approximations to dynamic programming for
		stochastic control problems with local transitions (CN435)
		B. V. Roy, J. N. Tsitsiklis
		Massachusetts Institute of Technology

17:05		Stable fitted reinforcement learning (CN361, Spotlight)
		G. J. Gordon
		Carnegie Mellon University

		Improving policies without measuring merits (CN441,
		Spotlight)
		P. Dayan, S. P. Singh
		Massachusetts Institute of Technology

17:10		Memory-based stochastic optimization (CN186)
		A. W. Moore, J. Schneider
		Carnegie Mellon University


THU NOV 30


		Oral Session 12:  Algorithms & Architectures

08:30		Statistical ideas for selecting network architectures
		(INVITED TALK)
		Brian Ripley
		Oxford University

09:00		SPERT-II:  A vector microprocessor system and its application 
		to large problems in backpropagation training (IM35)
		J. Wawrzynek, K. Asanovic, B. Kingsbury, J. Beck, D. Johnson, 
		N. Morgan
		University of California at Berkeley

09:20		Softassign verses softmax:  Benchmarks in combinatorial
		optimization (AA137)
		S. Gold, A. Rangarajan
		Yale University

09:40		A multiscale attentional framework for relaxation neural
		networks (AA445)
		D. I. Tsioutsias, E. Mjolsness
		Yale University


10:00-10:30	Break

		
10:30		Is learning the n-th thing any easier than learning the
		first? (AA17)
		S. Thrun
		University of Bonn

10:50		Using unlabeled data for supervised learning (AA316)
		G. Towell
		Siemens Corporate Research

11:10		Learning sparse perceptrons (AA352)
		J. C. Jackson, M. W. Craven
		University of Wisconsin-Madison

11:30		Does the wake-sleep algorithm learn good density
		estimators? (AA397)
		B. J. Frey, G. E. Hinton, P. Dayan
		University of Toronto


-------------------------------------------------------------------------------

TUE NOV 28
----------

19:30-22:30	Poster Session

		REMAP:  Recursive Estimation and Maximization of A
		posteriori Probabilities -- Application to transition-based
		connectionist speech recognition (AA94)
		Y. Konig, H. Bourlard, N. Morgan
		International Computer Science Institute

		Recurrent neural networks for missing or asynchronous data
		(AA238)
		Y. Bengio, F. Gingras
		Universite de Montreal

		Family discovery (AA311)
		S. M. Omohundro
		NEC Research Institute

		Discriminant adaptive nearest neighbor classification and
		regression (AA264)
		T. Hastie, R. Tibshirani
		Stanford University

		Clustering data through an analogy to the Potts model (AA30)
		M. Blatt, S. Wiseman, E. Domany
		The Weizmann Institute of Science

		Generalized learning vector quantization (AA36)
		A. Sato, K. Yamada
		NEC Corporation

		Stochastic hillclimbing as a baseline method for evaluating
		genetic algorithms (AA64)
		A. Juels, M. Wattenberg
		University of California, Berkeley

		Symplectic nonlinear component analysis (AA69)
		L. C. Parra
		Siemens Corporate Research

		A unified learning scheme:  Bayesian-Kullback coupling
		machine (AA72)
		L. Xu
		The Chinese University of Hong Kong

		Universal approximation and learning of trajectories using
		oscillators (AA77)
		P. Baldi, K. Hornik
		California Institute of Technology

		A smoothing regularizer for recurrent neural networks (AA78)
		L. Wu, J. Moody
		Oregon Graduate Institute

		A fast EM algorithm for latent-variable density models (AA133)
		C. M. Bishop, M. Svensen, C. K. I. Williams
		Aston University

		Factorial hidden markov models (AA139)
		Z. Ghahramani, M. I. Jordan
		Massachusetts Institute of Technology

		Boosting decision trees (AA173)
		H. Drucker, C. Cortes
		AT&T Bell Laboratories

		Exploiting tractable substructures in intractable networks
		(AA190)
		L. K. Saul, M. I. Jordan
		Massachusetts Institute of Technology

		Hierarchical recurrent neural networks for long-term
		dependencies (AA237)
		S. E. Hihi, Y. Bengio
		Universite de Montreal

		Human face detection in visual scenes (AP18)
		H. A. Rowley, S. Baluja, T. Kanade
		Carnegie Mellon University

		Improving committee diagnosis with resampling techniques
		(AP319)
		B. Parmanto, P. W. Munro, H. R. Doyle
		University of Pittsburgh

		Primitive manipulation learning with connectionism (AP74)
		Y. Matsuoka
		Massachusetts Institute of Technology

		Beating a defender in robotic soccer:  Memory-based learning
		of a continuous function (AP99)
		P. Stone, M. Veloso
		Carnegie Mellon University

		Visual gesture-based robot guidance with a modular neural
		system (AP120)
		E. Littmann, A. Drees, H. Ritter
		Bielefeld University

		A novel channel selection system in cochlear implants using
		artificial neural network (AP157)
		M. A. Jabri, R. J. Wang
		Sydney University

		Prediction of beta sheets in proteins (AP267)
		A. Krogh, S. K. Riis
		NORDITA

		A dynamical systems approach for a learnable autonomous
		robot (CN38)
		J. Tani, N. Fukumura
		Sony Computer Science Laboratory

		Parallel optimization of motion controllers via policy
		iteration (CN86)
		J. A. Coelho, R. Sitaramen, R. A. Grupen
		University of Massachusetts, Amherst

		Learning fine motion by Markov mixtures of experts (CN354)
		M. Meila, M. I. Jordan
		Massachusetts Institute of Technology

		Neural control for nonlinear dynamic systems (CN382)
		S. Yu, A. M. Annaswamy
		Massachusetts Institute of Technology

		The curse of dimensionality and human reading (CS192)
		G. L. Martin
		MCC

		Harmony networks do not work (CS456)
		R. Gourley
		Simon Fraser University

		Dynamics of attention as near saddle-node bifurcation
		behavior (CS252)
		H. Nakahara, K. Doya
		University of Tokyo

		Improved silicon cochlea using compatible lateral bipolar
		transistors (IM49)
		A. van Schaik, E. Fragniere, E. Vittoz
		Swiss Federal Institute of Technology

		Adaptive retina with center-surround receptive field (IM182)
		S. Liu, K. Boahen
		California Institute of Technology

		Neuron-MOS temporal winner search hardware for fully-parallel 
		data processing (IM292)
		T. Shibata, T. Nakai, T. Morimoto, R. Kaihara, T. Yamashita,
		T. Ohmi
		Tohoku University

		Analog VLSI processor implementing the continuous wavelet
		transform (IM337)
		R. T. Edwards, G. Cauwenberghs
		Johns Hopkins University

		A bound on the error of cross validation using the
		approximation and estimation rates, with consequences for
		the training-test split (LT147)
		M. Kearns
		AT&T Bell Laboratories

		Learning with ensembles:  How overfitting can be useful
		(LT262)
		P. Sollich, A. Krogh
		NORDITA

		On the computational power of noisy spiking neurons (LT1)
		W. Maass
		Technische Universitaet Graz

		A realizable learning task which exhibits overfitting (LT27)
		S. Boes
		Institute of Physical and Chemical Research (RIKEN)

		Stable dynamic parameter adaption (LT33)
		S. M. Rueger
		Technische Universitaet Berlin

		Estimating the Bayes risk from sample data (LT75)
		R. R. Snapp, T. Xu
		University of Vermont

		Recursive estimation of modular RBF networks (LT97)
		V. Kadirkamanathan, M. Kadirkamanathan
		University of Sheffield

		On neural networks with minimal weights (LT112)
		V. Bohossian, J. Bruck
		California Institute of Technology

		Modern analytic techniques to solve the dynamics of
		recurrent neural networks (LT146)
		A. C. C Coolen, S. N. Laughton, D. Sherrington
		University of Oxford

		Implementation issues in the Fourier transform algorithm
		(LT148)
		Y. Masour, S. Sahar
		Tel-Aviv University

		Generalisation of a class of continuous neural networks
		(LT184)
		J. Shawe-Taylor, J. Zhao
		University of London

		Gradient and Hamiltonian dynamics applied to learning in
		neural networks (LT213)
		J. W. Howse, C. T. Abdallah, G. L. Heileman
		The University of New Mexico

		Reorganization of somatosensory cortex after tactile
		training (NS333)
		R. S. Peterson, J. G. Taylor
		King's College London

		The role of activity in synaptic competition at the
		neuromuscular junction (NS42)
		S. R. H. Joseph, D. J. Willshaw
		Edinburgh University

		When is an integrate-and-fire neuron like a Poisson neuron?
		(NS177)
		C. F. Stevens, A. Zador
		The Salk Institute

		How perception guides production in birdsong learning (NS196)
		C. L. Fry
		University of California, San Diego

		The geometry of eye rotations and Listing's law (NS250)
		A. A. Handzel, T. Flash
		Weizmann Institute of Science

		A dynamical model of context dependencies for the 
		vestibulo-ocular reflex (NS452)
		O. J. M. D. Coenen, T. J. Sejnowski
		The Salk Institute

		Onset-based sound segmentation (SP103)
		L. S. Smith
		University of Stirling

		Laterally interconnected self-organizing maps in handwritten 
		digit rcognition (SP391)
		Y. Choe, J. Sirosh, R. Miikkulainen
		University of Texas, Austin

		A new learning algorithm for blind signal separation (SP34)
		S. Amari, A. Cichocki, H. H. Yang
		Institute of Physical and Chemical Research (RIKEN)

		Handwritten word recognition using contextual hybrid 
		RBF/hidden markov models (SP145)
		B. Lemarie, M. Gilloux, M. Leroux
		La Poste/SRTP

		A framework for nonrigid matching and correspondence
		(VS85)
		S. Pappu, S. Gold, A. Rangarajan
		Yale University

		Control of selective visual attention:  Modeling the "where"
		pathway (VS14)
		E. Niebur, C. Koch
		California Institute of Technology

		Unsupervised pixel-prediction (VS111)
		W. R. Softky
		NIDDK, NIH

		Learning to predict visibility and invisibility from
		occlusion events (VS308)
		J. A. Marshall, R. K. Alley, R. S. Hubbard
		University of North Carolina



WED NOV 29
----------

19:30-22:30	Poster Session

		Using pairs of data-points to define splits for decision
		trees (AA396)
		G. E. Hinton, M. Revow
		University of Toronto

		Regression with Gaussian processes (AA136)
		C. K. I. Williams and C. E. Rasmussen
		Aston University

		Pruning with generalization based weight saliencies:
		\gammaOBD, \gammaOBS (AA278)
		M. W. Pedersen, L. K. Hansen, J. Larsen
		Technical University of Denmark 

		Fast learning by bounding likelihoods in sigmoid belief
		networks (AA284)
		T. Jaakkola, L. K. Saul, M. I. Jordan
		Massachusetts Institute of Technology

		Generating accurate and diverse members of a neural-network
		ensemble (AA286)
		D. W. Opitz, J. W. Shavlik
		University of Wisconsin, Madison

		Improved Gaussian mixture density estimates using Bayesian
		penalty terms and network averaging (AA296)
		D. Ormoneit, V. Tresp
		Technische Universitaet Muenchen

		Explorations with the dynamic wave model (AA302)
		T. P. Rebotier, J. L. Elman
		University of California, San Diego

		The capacity of a bump (AA330)
		G. W. Flake
		Siemens Corporate Research

		Tempering backpropagation networks:  Not all weights are
		created equal (AA331)
		N. N. Schraudolph, T. J. Sejnowski
		The Salk Institute

		Investment learning with hierarchical PSOMs (AA347)
		J. Walter, H. Ritter
		Bielefeld University

		Learning long-term dependencies is not as difficult with
		NARX networks (AA394)
		T. Lin, B. G. Horne, P. Tino, C. L. Giles
		NEC Research Institute

		Constructive algorithms for hierarchical mixtures of
		experts (AA408)
		S. R. Waterhouse, A. J. Robinson
		Cambridge University

		An information-theoretic learning algorithm for neural
		network classification (AA417)
		D. Miller, A. Rao, K. Rose, A. Gersho
		University of California

		A practical Monte Carlo implementation of Bayesian learning
		(AA425)
		C. E. Rasmussen
		University of Toronto

		Cooperation in isolation:  An alternative view of a system
		of experts (AA439)
		S. Schaal, C. C. Atkeson
		Georgia Institute of Technology

		Finite state automata that recurrent cascade-correlation
		cannot represent (AA455)
		S. C. Kremer
		University of Alberta

		A neural network classifier for the I1000 OCR chip (AP221)
		J. C. Platt, T. P. Allen
		Synaptics

		Predictive Q-routing:  A memory-based reinforcement learning
		approach to adaptive traffic control (AP129)
		S. P. M. Choi, D. Yeung
		The Hong Kong University of Science and Technology

		Optimal portfolio management using adaptive dynamic
		programming (AP298)
		R. Neuneier
		Siemens AG

		Using the future to "sort out" the present:  Rankprop and
		multitask learning for medical risk evaluation (AP19)
		R. Caruana, S. Baluja, T. Mitchell
		Carnegie Mellon University

		Stock selection via nonlinear multi-factor models (AP329)
		A. U. Levin
		Wells Fargo Nikko Investment Advisors

		Experiments with neural networks for real time
		implementation of optimal control (AP369)
		P. Campbell, M. Dale, H. L. Ferra, A. Kowalczyk
		Telstra Research Laboratories

		High-speed airborne particle monitoring using artificial
		neural networks (AP402)
		A. Ferguson, T. Sabisch, P. Kaye, L. C. Dixon, H. Bolouri
		University of Hertfordshire

		High-performance job-shop scheduling with a time-delay
		TD(lambda) network (CN387)
		W. Zhang, T. G. Dietterich
		Oregon State University

		Stable fitted reinforcement learning (CN361)
		G. J. Gordon
		Carnegie Mellon University

		Improving policies without measuring merits (CN441)
		P. Dayan, S. P. Singh
		Massachusetts Institute of Technology

		Continuous-time TD learning and computation in the basal
		ganglia (CN126)
		K. Doya
		ATR Human Information Processing Research Laboratories

		Reinforcement learning by probability matching (CN443)
		P. N. Sabes, M. I. Jordan
		Massachusetts Institute of Technology

		Rapid quality estimation of neural network input
		representations (CS287)
		K. J. Cherkauer, J. W. Shavlik
		University of Wisconsin, Madison

		A model of auditory stream segmentation (CS449)
		S. L. McCabe, M. J. Denham
		University of Plymouth

		Silicon models for auditory scene analysis (IM378)
		J. Lazzaro, J. Wawrzynek
		University of California, Berkeley

		A visual smooth pursuit tracking chip (IM459)
		R. Etienne-Cummings, J. Van der Spiegel
		Southern Illinois University

		Model matching and SFMD computation (IM219)
		S. Rehfuss, D. Hammerstrom
		Oregon Graduate Institute

		Parallel analog VLSI architectures for computation of
		heading direction and time-to-contact (IM15)
		G. Indiveri, J. Kramer, C. Koch
		California Institute of Technology

		Optimization principles for the neural code (LT363)
		M. DeWeese, W. Bialek
		The Salk Institute

		Strong unimodality and efficient learning of \mu-Perceptron
		networks (LT375)
		M. Marchand, S. Hadjifaradji
		University of Ottawa

		Active learning in multilayer perceptrons (LT141)
		K. Fukumizu
		Ricoh Corp.

		There is no good squashing function for the square loss (LT271)
		P. Auer, M. Herbster, M. K. Warmuth
		University of California, Santa Cruz

		Adaptive gradient descent in on-line learning of multilayer
		networks (LT274)
		A. H. L. West, D. Saad
		University of Edinburgh

		An optimization approach to mappings (LT303)
		G. J. Goodhill, S. Finch
		The Salk Institute

		Quadratic-type Lyapunov functions for competitive neural
		networks with different time-scales (LT370)
		A. Meyer-Baese, F. Ohl, H. Scheich
		Technical University of Darmstadt

		Examples of learning curves from a modified VC-formalism
		(LT372)
		A. Kowalczyk, J. Szymanski, P. L. Bartlett, R. C. Williamson
		Telecom Australia Research Labs

		Bayesian methods for mixtures of experts (LT409)
		S. Waterhouse, D. Mackay, T. Robinson
		Cambridge University

		Some results on convergent unlearning algorithm (LT422)
		S. A. Semenov, I. B. Shuvalova
		Institute of Physics and Technology, Moscow

		Geometry of early stopping in linear networks (LT440)
		R. Dodier
		University of Colorado, Boulder

		Absence of cycles in symmetric neural networks (LT454)
		X. Wang, A. Jagota, F. Botelho, M. Garzon
		University of California, Los Angeles

		Temporal coding in the submillisecond range:  Model of barn
		owl auditory pathway (NS304)
		R. Kempter, W. Gerster, J. L. van Hemmen, H. Wagner
		Technische Universitaet Muenchen

		Cholinergic suppression of synaptic transmission may allow
		combination of associative feedback and self-organizing
		feedforward connections in the neocortex (NS317)
		M. E. Hasselmo, M. Cekic
		Harvard University

		A predictive switching model of cerebellar movement control
		(NS386)
		a. G. Barto, J. T. Buckingham, J. C. Houk
		University of Massachusetts, Amherst

		Independent component analysis of electroencephalographic
		data (NS429)
		S. Makeig, A. J. Bell, T. Jung, T. J. Sejnowski
		Naval Health Research Center

		A thalamocortical circuit for computing directional heading
		in the rat (NS436)
		H. T. Blair
		Yale University

		Plasticity of center-surround opponent receptive fields in
		real and artificial neural systems of vision (NS446)
		S. Yasui, T. Furukawa
		Kyushu Institute of Technology

		Selective attention for handwritten digit recognition (SP185)
		E. Alpaydin
		Bogazici University

		KODAK IMAGELINK OCR alphanumeric handprint module (SP242)
		A. Shustorovich, C. W. Thrasher
		Eastman Kodak Company

		The gamma MLP for speech phoneme recognition (SP383)
		S. Lawrence, A. C. Tsoi, A. D. Back
		University of Queensland

		A neural network model of 3D lightness perception (VS70)
		L. Pessoa, W. Ross
		Boston University

		Empirical entropy manipulation for real-world problems (VS176)
		P. Viola, N. N. Schraudolph, T. J. Sejnowski
		The Salk Institute

		Active gesture recognition using learned visual attention
		(VS419)
		T. Darrell, A. Pentland
		Massachusetts Institute of Technology

		SEEMORE:  A neurally-inspired approach to visual object
		recognition (VS423)
		B. W. Mel
		University of Southern California