icml 2008 论文列表Machine Learning, Proceedings of the Twenty-Fifth International Conference (ICML 2008), Helsinki, Finland, June 5-9, 2008. |
Laplace maximum margin Markov networks.
Efficient multiclass maximum margin clustering.
Estimating local optimums in EM algorithm over Gaussian mixture model.
Improved Nyström low-rank approximation and error analysis.
Predicting diverse subsets using structural SVMs.
A quasi-Newton approach to non-smooth convex optimization.
Preconditioned temporal difference learning.
Democratic approximation of lexicographic preference models.
Listwise approach to learning to rank: theory and algorithm.
Fully distributed EM for very large datasets.
Efficiently learning linear-linear exponential family predictive representations of state.
Deep learning via semi-supervised embedding.
Fast solvers and efficient implementations for distance metric learning.
On multi-view active learning and the combination with semi-supervised learning.
Graph transduction via alternating minimization.
Adaptive p-posterior mixture-model kernels for multiple instance learning.
Dirichlet component analysis: feature extraction for compositional data.
Manifold alignment using Procrustes analysis.
Sparse multiscale gaussian process regression.
Prediction with expert advice for the Brier game.
Extracting and composing robust features with denoising autoencoders.
Beam sampling for the infinite hidden Markov model.
Topologically-constrained latent variable models.
A semiparametric statistical approach to model-free policy evaluation.
Training restricted Boltzmann machines using approximations to the likelihood gradient.
nu-support vector machine as conditional value-at-risk minimization.
The many faces of optimism: a unifying approach.
Composite kernel learning.
Apprenticeship learning using linear programming.
A least squares formulation for canonical correlation analysis.
Discriminative parameter learning for Bayesian networks.
Metric embedding for kernel classification rules.
Detecting statistical interactions with additive groves of trees.
Tailoring density estimation via reproducing kernel moment matching.
The asymptotics of semi-supervised learning in discriminative probabilistic models.
An RKHS for multi-view learning and manifold co-regularization.
Sample-based learning and search with permanent and transient memories.
Expectation-maximization for sparse and non-negative PCA.
mStruct: a new admixture model for inference of population structure in light of both genetic admixing and allele mutations.
A generalization of Haussler's convolution kernel: mapping kernel.
Data spectroscopy: learning mixture models using eigenspaces of convolution operators.
SVM optimization: inverse dependence on training set size.
Multi-classification by categorical features via clustering.
Compressed sensing and Bayesian experimental design.
Inverting the Viterbi algorithm: an abstract framework for structure design.
Fast incremental proximity search in large graphs.
Accurate max-margin training for structured output spaces.
Bayesian probabilistic matrix factorization using Markov chain Monte Carlo.
On the quantitative analysis of deep belief networks.
Privacy-preserving reinforcement learning.
Robust matching and recognition using context-dependent kernels.
The Group-Lasso for generalized linear models: uniqueness of solutions and efficient algorithms.
Bi-level path following for cross validated solution of kernel quantile regression.
Closed-form supervised dimensionality reduction with generalized linear models.
The dynamic hierarchical Dirichlet process.
Online kernel selection for Bayesian reinforcement learning.
Bayesian multiple instance learning: automatic feature selection and inductive transfer.
Message-passing for graph-structured linear programs: proximal projections, convergence and rounding schemes.
Semi-supervised learning of compact document representations with deep networks.
Learning diverse rankings with multi-armed bandits.
Estimating labels from label proportions.
Multi-task compressive sensing with Dirichlet process priors.
Learning to learn implicit queries from gaze patterns.
An analysis of linear models, linear value-function approximation, and feature selection for reinforcement learning.
On the chance accuracies of large collections of classifiers.
A distance model for rhythms.
Learning dissimilarities by ranking: from SDP to QP.
The projectron: a bounded kernel-based Perceptron.
Cost-sensitive multi-class classification from probability estimates.
A decoupled approach to exemplar-based unsupervised learning.
Bayes optimal classification for decision trees.
On the hardness of finding symmetries in Markov decision processes.
Efficiently solving convex relaxations for MAP estimation.
Empirical Bernstein stopping.
An analysis of reinforcement learning with function approximation.
Rank minimization via online learning.
Automatic discovery and transfer of MAXQ hierarchies.
Nonextensive entropic kernels.
On-line discovery of temporal-difference networks.
A reproducing kernel Hilbert space framework for pairwise time series distances.
Uncorrelated multilinear principal component analysis through successive variance maximization.
Random classification noise defeats all convex potential boosters.
ManifoldBoost: stagewise function approximation for fully-, semi- and un-supervised learning.
Structure compilation: trading structure for features.
An asymptotic analysis of generative, discriminative, and pseudolikelihood estimators.
Pairwise constraint propagation by semidefinite programming for semi-supervised classification.
Knows what it knows: a framework for self-aware learning.
A worst-case comparison between temporal difference and residual gradient with linear function approximation.
Local likelihood modeling of temporal text streams.
Transfer of samples in batch reinforcement learning.
Classification using discriminative restricted Boltzmann machines.
Exploration scavenging.
Modeling interleaved hidden processes.
Query-level stability and generalization in learning to rank.
Fast estimation of first-order clause coverage through randomization and maximum likelihood.
The skew spectrum of graphs.
Space-indexed dynamic programming: learning to follow trajectories.
On partial optimality in multi-label MRFs.
Unsupervised rank aggregation with distance-based models.
ICA and ISA using Schweizer-Wolff measure of dependence.
Non-parametric policy gradients: a unified treatment of propositional and relational domains.
Large scale manifold transduction.
Efficient bandit algorithms for online multiclass prediction.
Causal modelling combining instantaneous and lagged effects: an identifiable model based on non-Gaussianity.
Discriminative structure and parameter learning for Markov logic networks.
A dual coordinate descent method for large-scale linear SVM.
Active kernel learning.
Statistical models for partial membership.
Modified MMI/MPE: a direct evaluation of the margin in speech recognition.
Grassmann discriminant analysis: a unifying view on subspace-based learning.
Boosting with incomplete information.
No-regret learning in convex games.
Localized multiple kernel learning.
Memory bounded inference in topic models.
Reinforcement learning in the presence of rare events.
Stopping conditions for exact computation of leave-one-out error in support vector machines.
Optimized cutting plane algorithm for support vector machines.
An HDP-HMM for systems with state persistence.
Training structural SVMs when exact inference is intractable.
Active reinforcement learning.
Polyhedral classifier for target detection: a case study: colorectal cancer.
Pointwise exact bootstrap distributions of cost curves.
Efficient projections onto the l1-ball for learning in high dimensions.
Confidence-weighted linear classification.
Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs.
Optimizing estimated loss reduction for active sampling in rank learning.
An object-oriented representation for efficient reinforcement learning.
Learning from incomplete data with infinite imputations.
Maximum likelihood rule ensembles.
Learning to classify with missing and corrupted features.
Hierarchical sampling for active learning.
Self-taught clustering.
Fast Gaussian process methods for point process intensity estimation.
A rate-distortion one-class model and its applications to clustering.
Stability of transductive regression algorithms.
Autonomous geometric precision error estimation in low-level computer vision tasks.
A unified architecture for natural language processing: deep neural networks with multitask learning.
Spectral clustering with inconsistent advice.
Learning for control from multiple demonstrations.
Training SVM with indefinite kernels.
Learning to sportscast: a test of grounded language acquisition.
Nearest hyperdisk methods for high-dimensional classification.
Fast nearest neighbor retrieval for bregman divergences.
Fast support vector machine training and classification on graphics processors.
An empirical evaluation of supervised learning in high dimensions.
Sparse Bayesian nonparametric regression.
Actively learning level-sets of composite functions.
Strategy evaluation in extensive games with importance sampling.
Nonnegative matrix factorization via rank-one downdate.
Multi-task learning for HIV therapy screening.
Multiple instance ranking.
Learning all optimal policies with multiple criteria.
Bolasso: model consistent Lasso estimation through the bootstrap.
Graph kernels between point clouds.
Hierarchical kernel stick-breaking process for multi-task image analysis.
Sequence kernels for predicting protein essentiality.
Gaussian process product models for nonparametric nonstationarity.