## Advances in Neural Information Processing Systems 13: Proceedings of the 2000 ConferenceThe annual conference on Neural Information Processing Systems (NIPS) is the flagship conference on neural computation. The conference is interdisciplinary, with contributions in algorithms, learning theory, cognitive science, neuroscience, vision, speech and signal processing, reinforcement learning and control, implementations, and diverse applications. Only about 30 percent of the papers submitted are accepted for presentation at NIPS, so the quality is exceptionally high. These proceedings contain all of the papers that were presented at the 2000 conference. |

### What people are saying - Write a review

We haven't found any reviews in the usual places.

### Contents

Who Does What? A Novel Algorithm to Determine Function Localization | 3 |

A Productive Systematic Framework for the Representation of Visual Structure | 10 |

The Interplay of Symbolic and Subsymbolic Processes in Anagram Problem | 17 |

HippocampallyDependent Consolidation in a Hierarchical Model ofNeocortex | 24 |

Position Variance Recurrence and Perceptual Learning | 31 |

The Use ofMDL to Select among Computational Models of Cognition | 39 |

In J Myung Mark A Pitt Shaobo Zhang and Vijay Balasubramanian | 45 |

The Early Word Catches the Weights | 52 |

Feature Selection for SVMs Jason Weston Sayan Mukherjee Olivier Chapelle | 668 |

On a Connection between Kernel PCA and Metric Multidimensional Scaling | 675 |

Using the Nystrom Method to Speed Up Kernel Machines | 682 |

Generalized Belief Propagation | 689 |

A GradientBased Boosting Algorithm for Regression Problems | 696 |

A Silicon Primitive for Competitive Learning | 713 |

Homeostasis in a Silicon Integrate and Fire Neuron | 727 |

Fourlegged Walking Gait Control Using a Neuromorphic Chip Interfaced to | 741 |

Structure Learning in Human Causal Induction | 59 |

Adaptive Object Representation with HierarchicallyDistributed Memory Sites | 66 |

What Can a Single Neuron Compute? | 75 |

Stability and Noise in Biochemical Switches William Bialek | 103 |

A New Model of Spatial Representation in Multimodal Brain Areas | 119 |

Finding the Key to a Synapse Thomas Natschlager and Wolfgang Maass | 138 |

SpikeTimingDependent Learning for Oscillatory Networks | 152 |

Natural Sound Statistics and Divisive Normalization in the Auditory System | 166 |

Whence Sparseness? Carl van Vreeswijk | 180 |

Algorithmic Stability and Generalization Performance | 196 |

From Margin to Sparsity Thore Graepel Ralf Herbrich and Robert C Williamson | 210 |

Why SVMs work | 224 |

Second Order Approximations for Probability Models | 238 |

A Support Vector Method for Clustering | 367 |

A Variational MeanField Theory for Sigmoidal Belief Networks | 374 |

Direct Classification with Indirect Data Timothy X Brown | 381 |

Model Complexity Goodness of Fit and Diminishing Returns | 388 |

A Linear Programming Approach to Novelty Detection | 395 |

Incremental and Decremental Support Vector Machine Learning | 409 |

Gaussianization Scott Saobing Chen and Ramesh A Gopinath | 423 |

Improved Output Coding for Classification Using Continuous Relaxation | 437 |

Explaining Away in Weight Space Peter Dayan and Sham Kakade | 451 |

Hightemperature Expansions for Learning Models of Nonnegative Data | 465 |

A StructureBased Approach | 479 |

Sequentially Fitting Inclusive Trees for Inference in NoisyOR Networks | 493 |

Propagation Algorithms for Variational Bayesian Learning | 507 |

NBody Problems in Statistical Learning | 523 |

Sparse Kernel Principal Component Analysis Michael E Tipping | 633 |

Data Clustering by Markovian Relaxation and the Information Bottleneck | 640 |

Active Learning for Parameter Estimation in Bayesian Networks | 647 |

Mixtures of Gaussian Processes VolkerTresp | 654 |

TreeBased Modeling and Estimation of Gaussian Processes on Graphs with | 661 |

Speech Denoising and Dereverberation Using Probabilistic Models | 758 |

Learning Joint Statistical Models for AudioVisual Fusion and Segregation | 772 |

HigherOrder Statistical Properties Arising from the NonStationarity of Natural | 786 |

Minimum Bayes Error Feature Selection for Continuous Speech Recognition | 800 |

A Linear Operator for Measuring Synchronization of Video Facial | 814 |

A New Descriptor for Shape Matching and Object Recognition | 831 |

Emergence of Movement Sensitive Neurons Properties by Learning a Sparse | 838 |

Regularities in Scene Statistics which Enable | 845 |

A Markov Chain Monte Carlo Approach | 852 |

Keeping Flexible Active Contours on Track using Metropolis Updates | 859 |

Color Opponency Constitutes a Sparse Representation for the Chromatic | 866 |

Learning Segmentation by Random Walks Marina MeilS and Jianbo Shi | 873 |

Partially Observable SDE Models for Image Sequence Recognition Tasks | 880 |

Learning Sparse Image Codes using a Wavelet Pyramid Architecture | 887 |

Learning and Tracking Cyclic Human Motion | 894 |

Redundancy and Dimensionality Reduction in SparseDistributed | 901 |

Ratecoded Restricted Boltzmann Machines for Face Recognition | 908 |

Linking Psychophysics and Biophysics | 915 |

From Mixtures of Mixtures to Adaptive Transform Coding | 925 |

A Neural Probabilistic Language Model | 932 |

A Comparison of Image Processing Techniques for Visual Speech Recognition | 939 |

Recognizing Handwritten Digits Using Hierarchical Products of Experts | 953 |

Probabilistic Semantic Video Indexing | 967 |

Learning Switching Linear Models of Human Motion | 981 |

The Use of Classifiers in Sequential Inference Vasin Punyakanok and Dan Roth | 995 |

Bayesian Video Shot Segmentation Nuno Vasconcelos and Andrew Lippman | 1009 |

Exact Solutions to TimeDependent MDPs Justin A Boyan and Michael L Littman | 1026 |

Reinforcement Learning with Function Approximation Converges to a Region | 1040 |

Automated State Abstraction for Options using the UTree Algorithm | 1054 |

Balancing Multiple Sources of Reward in Reinforcement Learning | 1082 |

### Common terms and phrases

active Advances in Neural analysis applied approach approximation basis functions Bayesian Boltzmann machine bound cells classifier clustering complexity components convergence correlation corresponding covariance covariance matrix data set defined denotes density dimensional distribution dynamics eigenvalues equation error estimate example feature space Figure filter frequency Gaussian processes given gradient hidden units hippocampus IEEE independent independent component analysis Information Processing Systems input iterative kernel kernel PCA layer learning algorithm linear Machine Learning margin matrix maximal method minimize mutual information Neural Information Processing neural networks neurons nodes noise nonlinear normalized obtained optimal output parameters patterns perceptron performance pixels points posterior prediction prior probabilistic probability problem random receptive fields recognition representation sample sequence shown signal solution speech spike statistical stimulus structure Support Vector Machines synapses Theorem theory training data training set transform unsupervised learning update values variance visual weights