Advances in Neural Information Processing Systems 17: Proceedings of the 2004 ConferenceLawrence K. Saul, Yair Weiss, Léon Bottou Papers presented at NIPS, the flagship meeting on neural computation, held in December 2004 in Vancouver.The annual Neural Information Processing Systems (NIPS) conference is the flagship meeting on neural computation. It draws a diverse group of attendees--physicists, neuroscientists, mathematicians, statisticians, and computer scientists. The presentations are interdisciplinary, with contributions in algorithms, learning theory, cognitive science, neuroscience, brain imaging, vision, speech and signal processing, reinforcement learning and control, emerging technologies, and applications. Only twenty-five percent of the papers submitted are accepted for presentation at NIPS, so the quality is exceptionally high. This volume contains the papers presented at the December, 2004 conference, held in Vancouver. |
Contents
A Large Deviation Bound for the Area Under the ROC Curve | 9 |
Harmonising Chorales by Probabilistic Inference MORAY ALLAN | 25 |
A Direct Formulation for Sparse PCA Using Semidefinite | 41 |
An application to object | 57 |
Computing regularization paths for learning multiple kernels | 73 |
Breaking SVM Complexity with CrossTraining GÖKHAN | 81 |
LargeScale Prediction of Disulphide Bond Connectivity PIERRE | 97 |
Bayesian Clustering of NonStationary Data | 105 |
Semisupervised Learning with Penalized Probabilistic Clustering | 849 |
Methods for Estimating the Computational Power | 865 |
PACBayes Learning of Conjunctions and Classification | 881 |
Linear Multilayer Independent Component Analysis for Large | 897 |
Multiple Relational Embedding ROLAND MEMISEVIC and GEOFFREY | 913 |
Classification | 929 |
Validity Estimates for Loopy Belief Propagation on Binary | 945 |
Optimal subgraphical models MUKUND NARASIMHAN and JEFF | 961 |
Maximising Sensitivity in a Spiking Network ANTHONY J BELL | 121 |
Whos In the Picture TAMARA L BERG ALEXANDER C BERG | 137 |
A Second Order Cone programming Formulation for Classifying | 153 |
Responding to Modalities with Different Latencies FREDRIK | 169 |
Hierarchical Distributed Representations for Statistical Language | 185 |
A Computational Theory | 201 |
Dependent Gaussian Processes PHILLIP BOYLE and MARCUS FREAN | 217 |
Incremental Algorithms for Hierarchical Classification NICOLÒ | 233 |
SubMicrowatt Analog VLSI Support Vector Machine for Pattern | 249 |
Using Machine Learning to Break Visual Human Interaction | 265 |
Modeling Conversational Dynamics as a MixedMemory Markov | 281 |
Distributed Information Regularization on Graphs ADRIAN | 297 |
A Latent | 313 |
Semigroup Kernels on Finite Sets MARCO CUTURI | 329 |
SelfBounded Learning | 345 |
Triangle Fixing Algorithms for the Metric Nearness Problem | 361 |
Sparse Coding of Natural Images Using an Overcomplete Set | 377 |
Seeing through water ALEXEI EFROS University of Oxford VOLKAN | 393 |
ExplorationExploitation Tradeoffs for Experts Algorithms | 409 |
Learning HyperFeatures for Visual Identification ANDRAS | 425 |
OnChip Compensation of DeviceMismatch Effects in Analog | 441 |
A Hidden Markov Model for de Novo Peptide Sequencing BERND | 457 |
Joint Probabilistic Curve Clustering and Alignment SCOTT | 473 |
InstanceBased Relevance Feedback for Image Retrieval GIORGIO | 489 |
Hierarchical Clustering of a Mixture Model JACOB GOLDBERGER | 505 |
The Cascade SVM HANS | 521 |
Integrating Topics and Syntax THOMAS L GRIFFITHS Massachusetts | 537 |
characteristic propagation | 553 |
An Auditory Paradigm for BrainComputer Interfaces N JEREMY | 569 |
HOFSTOETTER MANUEL GIL KYNAN ENG GIACOMO INDIVERI | 577 |
Unsupervised Variational Bayesian Learning of Nonlinear Models | 593 |
Message Errors in Belief Propagation ALEXANDER T IHLER JOHN | 609 |
A Cost Function for Clustering | 625 |
Online Bounds for Bayesian Algorithms SHAM M KAKADE | 641 |
Generalization Error and Algorithmic Convergence of Median | 657 |
Face Detection Efficient and Rank Deficient WOLF KIENZLE | 673 |
Synchronization of neural networks by mutual learning and | 689 |
Optimal Aggregation of Classifiers and Boosting Maps | 705 |
On SemiSupervised Classification BALAJI KRISHNAPURAM | 721 |
Methods Towards Invasive Human Brain Computer Interfaces | 737 |
Semisupervised Learning via Gaussian Processes NEIL D | 753 |
Rate and Phasecoded Autoassociative Memory MÁTÉ LENGYEL | 769 |
Planning for Markov Decision Processes with Sparse Stochasticity | 785 |
Adaptive Discriminative Generative Model and Its Applications | 801 |
Multiple Alignment of Continuous Time Series JENNIFER | 817 |
Mistake Bounds for Maximum Entropy Discrimination PHILIP M | 833 |
Stable adaptive control with online learning ANDREW Y | 977 |
A Harmonic Excitation StateSpace Approach to Blind Separation | 993 |
Discrete profile alignment via constrained information bottleneck | 1009 |
Logconcavity Results on Gaussian Process Methods | 1025 |
Modeling Nonlinear Dependencies in Natural Images using | 1041 |
Efficient OutofSample Extension of DominantSet Clusters | 1057 |
Active Learning for Anomaly and RareCategory Detection | 1073 |
New Criteria and a New Algorithm for Learning in MultiAgent | 1089 |
Chemosensory Processing in a Spiking Model of the Olfactory | 1105 |
An Information Maximization Model of Eye Movements LAURA | 1121 |
Coarticulation in Markov Decision Processes KHASHAYAR | 1137 |
Following Curved Regularized Optimization Solution Paths | 1153 |
Outlier Detection with Oneclass Kernel Fisher Discriminants | 1169 |
SemiMarkov Conditional Random Fields for Information | 1185 |
Edge of Chaos Computation in MixedMode VLSI A Hard | 1201 |
Assignment of Multiplicative Mixtures in Natural Images ODELIA | 1217 |
RealTime Pitch Determination of One or More Voices | 1233 |
Resolving Perceptual Aliasing In The Presence Of Noisy Sensors | 1249 |
Dynamic Bayesian Networks for BrainComputer Interfaces | 1265 |
Intrinsically Motivated Reinforcement Learning SATINDER SINGH | 1281 |
Learning Syntactic Patterns for Automatic Hypernym Discovery | 1297 |
Using the Equivalent Kernel to Understand Gaussian Process | 1313 |
MaximumMargin Matrix Factorization NATHAN SREBRO University | 1329 |
Fast Rates to Bayes for Kernel Machines INGO STEINWART | 1345 |
Modelling Uncertainty in the Game of Go DAVID H STERN | 1353 |
Distributed Occlusion Reasoning for Tracking with Nonparametric | 1369 |
Hierarchical Dirichlet | 1385 |
Contextual Models for Object Detection Using Boosted Random | 1401 |
Synergies between Intrinsic and Synaptic Plasticity in Individual | 1417 |
Supervised Graph Inference JEANPHILIPPE VERT Ecole des Mines | 1433 |
InstanceSpecific Bayesian Model Averaging for Classification | 1449 |
Identifying ProteinProtein Interaction Sites on a GenomeWide | 1465 |
Exponential Family Harmoniums with an Application | 1481 |
The Variational Ising Classifier VIC Algorithm for Coherently | 1497 |
lonorm Minimization for Basis Selection DAVID P WIPF | 1513 |
Efficient Kernel Discriminant Analysis via QR Decomposition | 1529 |
Using Random Forests in the Structured Language Model PENG | 1545 |
Efficient Kernel Machines Using the Improved Fast Gauss | 1561 |
Inference Attention and Decision in a Bayesian Neural | 1577 |
The Convergence of Contrastive Divergences ALAN YUILLE UCLA | 1593 |
Probabilistic Computation in Spiking Populations RICHARD | 1609 |
Classsize Independent Generalization Analsysis of Some | 1625 |
Nonparametric Transforms of Graph Kernels for SemiSupervised | 1641 |
1657 | |
Common terms and phrases
ADABOOST Advances in Neural analysis applied approach approximation assume average Bayesian belief propagation bound classifier clustering components Computer Computer Vision constraints convergence corresponding cysteines datasets defined denote density dimensionality distance distribution eigenvectors EM algorithm embedding entropy error estimate experiments Figure Gaussian given graph Hidden Markov Model IEEE Information Processing Systems input Isomap iteration kernel labels learning algorithms linear logistic regression Machine Learning manifold margin Markov matrix method minimize mixture Neural Information Processing neural networks neuron node noise nonlinear object observed obtained optimal output parameters patches pattern perceptron performance pixels points prediction probabilistic probability problem proposed random regression representation sample segmentation selection semi-supervised semi-supervised learning sequence signal solution space spectral clustering spike statistical structure subset support vector support vector machines synaptic Theorem training set update variables variance weight