ICML 2008 - PDF Preprint Papers

ICML 2008 Workshop on Statistical Network Analysis

Teaser Browsing | Click Here

Title Browsing | Click Here

Wall of Fame | Most Viewed ICML-2008 Paper

click to vote

ICML
2008
IEEE

278views Machine Learning» more ICML 2008»

Discriminative parameter learning for Bayesian networks

15 years 7 days ago

Download www.cs.unb.ca

Bayesian network classifiers have been widely used for classification problems. Given a fixed Bayesian network structure, parameters learning can take two different approaches: ge...

Jiang Su, Harry Zhang, Charles X. Ling, Stan Matwi...

claim paper

Read More »

Disclaimer and Copyright Notice

Sciweavers respects the rights of all copyright holders and in this regard, authors are only allowed to share a link to their preprint paper on their own website. Every contribution is associated with a desciptive image. It is the sole responsibility of the authors to ensure that their posted image is not copyright infringing. This service is compliant with IEEE copyright.

Id	Views	Title	Status

1	278	Discriminative parameter learning for Bayesian networks	claim
2	241	Discriminative structure and parameter learning for Markov logic networks	claim
3	233	Bayesian multiple instance learning: automatic feature selection and inductive transfer	claim
4	229	Classification using discriminative restricted Boltzmann machines	claim
5	217	Active kernel learning	claim
6	211	ICA and ISA using Schweizer-Wolff measure of dependence	claim
7	198	On multi-view active learning and the combination with semi-supervised learning	claim
8	197	Laplace maximum margin Markov networks	claim
9	195	Estimating local optimums in EM algorithm over Gaussian mixture model	claim
10	195	Optimizing estimated loss reduction for active sampling in rank learning	claim
11	190	The asymptotics of semi-supervised learning in discriminative probabilistic models	claim
12	188	Inverting the Viterbi algorithm: an abstract framework for structure design	claim
13	179	Efficiently solving convex relaxations for MAP estimation	claim
14	177	Accurate max-margin training for structured output spaces	claim
15	177	Sparse multiscale gaussian process regression	claim
16	176	Bayesian probabilistic matrix factorization using Markov chain Monte Carlo	claim
17	176	Cost-sensitive multi-class classification from probability estimates	claim
18	173	Adaptive p-posterior mixture-model kernels for multiple instance learning	claim
19	171	An asymptotic analysis of generative, discriminative, and pseudolikelihood estimators	claim
20	171	Modeling interleaved hidden processes	claim
21	170	On the quantitative analysis of deep belief networks	claim
22	169	Large scale manifold transduction	claim
23	169	Robust matching and recognition using context-dependent kernels	claim
24	169	Listwise approach to learning to rank: theory and algorithm	claim
25	166	ManifoldBoost: stagewise function approximation for fully-, semi- and un-supervised learning	claim
26	165	A worst-case comparison between temporal difference and residual gradient with linear function approximation	claim
27	162	Automatic discovery and transfer of MAXQ hierarchies	claim
28	162	Efficient bandit algorithms for online multiclass prediction	claim
29	161	Unsupervised rank aggregation with distance-based models	claim
30	160	Localized multiple kernel learning	claim
31	160	Bi-level path following for cross validated solution of kernel quantile regression	claim
32	157	Efficiently learning linear-linear exponential family predictive representations of state	claim
33	156	Gaussian process product models for nonparametric nonstationarity	claim
34	155	A least squares formulation for canonical correlation analysis	claim
35	155	Extracting and composing robust features with denoising autoencoders	claim
36	154	Beam sampling for the infinite hidden Markov model	claim
37	153	Confidence-weighted linear classification	claim
38	152	Modified MMI/MPE: a direct evaluation of the margin in speech recognition	claim
39	151	On-line discovery of temporal-difference networks	claim
40	150	Multi-task compressive sensing with Dirichlet process priors	claim
41	150	A generalization of Haussler's convolution kernel: mapping kernel	claim
42	150	An analysis of reinforcement learning with function approximation	claim
43	149	Topologically-constrained latent variable models
44	148	Semi-supervised learning of compact document representations with deep networks	claim
45	147	Apprenticeship learning using linear programming	claim
46	147	Graph transduction via alternating minimization	claim
47	146	Fast incremental proximity search in large graphs	claim
48	146	Causal modelling combining instantaneous and lagged effects: an identifiable model based on non-Gaussianity	claim
49	146	Boosting with incomplete information	claim
50	145	Deep learning via semi-supervised embedding	claim
51	144	An HDP-HMM for systems with state persistence	claim
52	143	Training structural SVMs when exact inference is intractable	claim
53	142	Learning to sportscast: a test of grounded language acquisition	claim
54	142	Composite kernel learning	claim
55	142	Fast solvers and efficient implementations for distance metric learning	claim
56	141	Uncorrelated multilinear principal component analysis through successive variance maximization	claim
57	139	Fast estimation of first-order clause coverage through randomization and maximum likelihood	claim
58	138	Nearest hyperdisk methods for high-dimensional classification	claim
59	138	Fully distributed EM for very large datasets	claim
60	136	Structure compilation: trading structure for features	claim
61	136	Training SVM with indefinite kernels	claim
62	135	Reinforcement learning with limited reinforcement: using Bayes risk for active learning in POMDPs	claim
63	134	Fast nearest neighbor retrieval for bregman divergences	claim
64	134	Dirichlet component analysis: feature extraction for compositional data	claim
65	134	A dual coordinate descent method for large-scale linear SVM	claim
66	133	Multi-classification by categorical features via clustering	claim
67	133	Space-indexed dynamic programming: learning to follow trajectories	claim
68	133	An analysis of linear models, linear value-function approximation, and feature selection for reinforcement learning	claim
69	132	Metric embedding for kernel classification rules	claim
70	131	Fast Gaussian process methods for point process intensity estimation	claim
71	130	Nonextensive entropic kernels	claim
72	130	Statistical models for partial membership	claim
73	130	A reproducing kernel Hilbert space framework for pairwise time series distances	claim
74	130	Compressed sensing and Bayesian experimental design	claim
75	130	Optimized cutting plane algorithm for support vector machines	claim
76	129	Multiple instance ranking	claim
77	129	Fast support vector machine training and classification on graphics processors	claim
78	129	Closed-form supervised dimensionality reduction with generalized linear models	claim
79	129	A decoupled approach to exemplar-based unsupervised learning	claim
80	127	Memory bounded inference in topic models	claim
81	127	On partial optimality in multi-label MRFs	claim
82	127	Learning to learn implicit queries from gaze patterns	claim
83	126	Strategy evaluation in extensive games with importance sampling	claim
84	126	An empirical evaluation of supervised learning in high dimensions	claim
85	124	Graph kernels between point clouds	claim
86	123	Bolasso: model consistent Lasso estimation through the bootstrap	claim
87	123	An object-oriented representation for efficient reinforcement learning	claim
88	122	Detecting statistical interactions with additive groves of trees	claim
89	122	Reinforcement learning in the presence of rare events	claim
90	120	Exploration scavenging	claim
91	119	Manifold alignment using Procrustes analysis	claim
92	119	Expectation-maximization for sparse and non-negative PCA	claim
93	119	Message-passing for graph-structured linear programs: proximal projections, convergence and rounding schemes	claim
94	118	Learning to classify with missing and corrupted features	claim
95	118	Knows what it knows: a framework for self-aware learning	claim
96	118	The dynamic hierarchical Dirichlet process	claim
97	117	A unified architecture for natural language processing: deep neural networks with multitask learning	claim
98	117	Sample-based learning and search with permanent and transient memories	claim
99	116	Predicting diverse subsets using structural SVMs	claim
100	116	Learning for control from multiple demonstrations	claim
101	115	Data spectroscopy: learning mixture models using eigenspaces of convolution operators	claim
102	115	On the hardness of finding symmetries in Markov decision processes	claim
103	114	Estimating labels from label proportions	claim
104	114	SVM optimization: inverse dependence on training set size	claim
105	113	Tailoring density estimation via reproducing kernel moment matching	claim
106	113	Democratic approximation of lexicographic preference models	claim
107	113	Rank minimization via online learning	claim
108	113	Query-level stability and generalization in learning to rank	claim
109	113	Training restricted Boltzmann machines using approximations to the likelihood gradient	claim
110	112	On the chance accuracies of large collections of classifiers	claim
111	111	Empirical Bernstein stopping	claim
112	110	The projectron: a bounded kernel-based Perceptron	claim
113	110	Non-parametric policy gradients: a unified treatment of propositional and relational domains	claim
114	110	Hierarchical sampling for active learning	claim
115	109	Learning diverse rankings with multi-armed bandits	claim
116	105	Learning all optimal policies with multiple criteria	claim
117	105	No-regret learning in convex games	claim
118	100	Stability of transductive regression algorithms	claim
119	100	The skew spectrum of graphs	claim
120	99	Random classification noise defeats all convex potential boosters	claim
121	98	A distance model for rhythms	claim
122	97	An RKHS for multi-view learning and manifold co-regularization	claim
123	92	Actively learning level-sets of composite functions	claim
124	90	Sequence kernels for predicting protein essentiality	claim
125	85	Stopping conditions for exact computation of leave-one-out error in support vector machines	claim
126	83	A rate-distortion one-class model and its applications to clustering	claim

Sciweavers

Explore & Download

Productivity Tools

Document Tools

Image Tools

Sciweavers