## Carla E. Brodley

*Proceedings of the 21st International Conference on Machine Learning*

ICML, 2004.

@proceedings{ICML-2004, address = "Banff, Alberta, Canada", editor = "Carla E. Brodley", publisher = "{ACM}", series = "{ACM International Conference Proceeding Series}", title = "{Proceedings of the 21st International Conference on Machine Learning}", volume = 69, year = 2004, }

### Contents (117 items)

- ICML-2004-AgarwalT #3d #learning
- Learning to track 3D human motion from silhouettes (AA, BT).
- ICML-2004-AhnCO #algorithm #multi
- A multiplicative up-propagation algorithm (JHA, SC, JHO).
- ICML-2004-AltunHS #classification #process #sequence
- Gaussian process classification for segmenting and annotating sequences (YA, TH, AJS).
- ICML-2004-AppiceCRF #multi #problem
- Redundant feature elimination for multi-class problems (AA, MC, SR, PAF).
- ICML-2004-BachLJ #algorithm #kernel #learning #multi
- Multiple kernel learning, conic duality, and the SMO algorithm (FRB, GRGL, MIJ).
- ICML-2004-BahamondeBDQLCAG #case study #learning #set
- Feature subset selection for learning preferences: a case study (AB, GFB, JD, JRQ, OL, JJdC, JA, FG).
- ICML-2004-BanerjeeDGM #analysis #estimation #exponential #product line
- An information theoretic analysis of maximum likelihood mixture estimation for exponential families (AB, ISD, JG, SM).
- ICML-2004-BasilicoH #collaboration
- Unifying collaborative and content-based filtering (JB, TH).
- ICML-2004-BaskiotisS #approach
- C4.5 competence map: a phase transition-inspired approach (NB, MS).
- ICML-2004-BilenkoBM #clustering #constraints #learning #metric
- Integrating constraints and metric learning in semi-supervised clustering (MB, SB, RJM).
- ICML-2004-BleiJ #process
- Variational methods for the Dirichlet process (DMB, MIJ).
- ICML-2004-BlumLRR #learning #random #using
- Semi-supervised learning using randomized mincuts (AB, JDL, MRR, RR).
- ICML-2004-BohteBG #classification #parametricity #polynomial
- Nonparametric classification with polynomial MPMC cascades (SMB, MB, GZG).
- ICML-2004-Bouckaert #classification #learning
- Estimating replicability of classifier learning experiments (RRB).
- ICML-2004-BrefeldS #learning
- Co-EM support vector learning (UB, TS).
- ICML-2004-Brinker #learning #ranking
- Active learning of label ranking functions (KB).
- ICML-2004-CaruanaNCK #library #modelling
- Ensemble selection from libraries of models (RC, ANM, GC, AK).
- ICML-2004-CastilloW #case study #comparative #learning #multi
- A comparative study on methods for reducing myopia of hill-climbing search in multirelational learning (LPC, SW).
- ICML-2004-ChangY #adaptation #clustering #linear #metric
- Locally linear metric adaptation for semi-supervised clustering (HC, DYY).
- ICML-2004-ChuGW #predict #visual notation
- A graphical model for protein secondary structure prediction (WC, ZG, DLW).
- ICML-2004-ClimerZ #approach #clustering
- Take a walk and cluster genes: a TSP-based approach to optimal rearrangement clustering (SC, WZ).
- ICML-2004-CollobertB
- Links between perceptrons, MLPs and SVMs (RC, SB).
- ICML-2004-ConitzerS #bound #communication #complexity #game studies #learning
- Communication complexity as a lower bound for learning in games (VC, TS).
- ICML-2004-CortesM #kernel
- Distribution kernels based on moments of counts (CC, MM).
- ICML-2004-CrammerC #optimisation
- A needle in a haystack: local one-class optimization (KC, GC).
- ICML-2004-DekelKS #classification #scalability
- Large margin hierarchical classification (OD, JK, YS).
- ICML-2004-DietterichAB #random
- Training conditional random fields via gradient tree boosting (TGD, AA, YB).
- ICML-2004-DingH #clustering
- Linearized cluster assignment via spectral ordering (CHQD, XH).
- ICML-2004-DingH04a #analysis #clustering #component
- K-means clustering via principal component analysis (CHQD, XH).
- ICML-2004-DSouzaVS
- The Bayesian backfitting relevance vector machine (AD, SV, SS).
- ICML-2004-EliazarP #learning #mobile #modelling #probability
- Learning probabilistic motion models for mobile robots (AIE, RP).
- ICML-2004-EsmeirM #algorithm #induction
- Lookahead-based algorithms for anytime induction of decision trees (SE, SM).
- ICML-2004-EspositoS #analysis #classification #monte carlo
- A Monte Carlo analysis of ensemble classification (RE, LS).
- ICML-2004-FernB #clustering #graph #problem
- Solving cluster ensemble problems by bipartite graph partitioning (XZF, CEB).
- ICML-2004-FernG #relational #reliability
- Relational sequential inference with reliable observations (AF, RG).
- ICML-2004-FerriFH #classification
- Delegating classifiers (CF, PAF, JHO).
- ICML-2004-Forman #classification #feature model #multi
- A pitfall and solution in multi-class feature selection for text classification (GF).
- ICML-2004-FrankK #multi #problem
- Ensembles of nested dichotomies for multi-class problems (EF, SK).
- ICML-2004-FungDBR #algorithm #kernel #performance #using
- A fast iterative algorithm for fisher discriminant using heterogeneous kernels (GF, MD, JB, RBR).
- ICML-2004-GabrilovichM #categorisation #feature model #using
- Text categorization with many redundant features: using aggressive feature selection to make SVMs competitive with C4.5 (EG, SM).
- ICML-2004-GaoWLC #approach #categorisation #learning #multi #robust
- A MFoM learning approach to robust multiclass multi-label text categorization (SG, WW, CHL, TSC).
- ICML-2004-Gilad-BachrachNT #algorithm #feature model
- Margin based feature selection — theory and algorithms (RGB, AN, NT).
- ICML-2004-GoldenbergM #learning #scalability
- Tractable learning of large Bayes net structures from sparse data (AG, AWM).
- ICML-2004-GramacyLM #parametricity #process
- Parameter space exploration with Gaussian process trees (RBG, HKHL, WGM).
- ICML-2004-GrossmanD #classification #learning #network
- Learning Bayesian network classifiers by maximizing conditional likelihood (DG, PMD).
- ICML-2004-HamLMS #kernel #reduction
- A kernel view of the dimensionality reduction of manifolds (JH, DDL, SM, BS).
- ICML-2004-HardinTA #feature model #linear
- A theoretical characterization of linear SVM-based feature selection (DPH, IT, CFA).
- ICML-2004-HerschtalR #optimisation #using
- Optimising area under the ROC curve using gradient descent (AH, BR).
- ICML-2004-HertzBW #clustering #distance
- Boosting margin based distance functions for clustering (TH, ABH, DW).
- ICML-2004-HuangYKL #classification #learning #scalability
- Learning large margin classifiers locally and globally (KH, HY, IK, MRL).
- ICML-2004-JakulinB #interactive #testing
- Testing the significance of attribute interactions (AJ, IB).
- ICML-2004-JamesS #learning #predict
- Learning and discovery of predictive state representations in dynamical systems with reset (MRJ, SPS).
- ICML-2004-JanodetNSS #grammar inference
- Boosting grammatical inference with confidence oracles (JCJ, RN, MS, HMS).
- ICML-2004-Jebara #kernel #multi
- Multi-task feature and kernel selection for SVMs (TJ).
- ICML-2004-JenkinsM #reduction
- A spatio-temporal extension to Isomap nonlinear dimension reduction (OCJ, MJM).
- ICML-2004-JinL #induction #robust
- Robust feature induction for support vector machines (RJ, HL).
- ICML-2004-KashimaT #algorithm #graph #kernel #learning #sequence
- Kernel-based discriminative learning algorithms for labeling sequences, trees, and graphs (HK, YT).
- ICML-2004-KerstingOR #relational
- Bellman goes relational (KK, MvO, LDR).
- ICML-2004-KimK #feature model
- Gradient LASSO for feature selection (YK, JK).
- ICML-2004-KokV
- Sparse cooperative Q-learning (JRK, NAV).
- ICML-2004-KoppelS #classification #problem #verification
- Authorship verification as a one-class classification problem (MK, JS).
- ICML-2004-KrauseS
- Leveraging the margin more carefully (NK, YS).
- ICML-2004-LaffertyZL #clique #kernel #random #representation
- Kernel conditional random fields: representation and clique selection (JDL, XZ, YL).
- ICML-2004-LawrenceP #learning
- Learning to learn with the informative vector machine (NDL, JCP).
- ICML-2004-LebanonL #classification #multi
- Hyperplane margin classifiers on the multinomial manifold (GL, JDL).
- ICML-2004-LeeWZB #perspective #probability
- Probabilistic tangent subspace: a unified view (JL, JW, CZ, ZB).
- ICML-2004-LiMO #category theory #clustering
- Entropy-based criterion in categorical clustering (TL, SM, MO).
- ICML-2004-LingYWZ #low cost
- Decision trees with minimal costs (CXL, QY, JW, SZ).
- ICML-2004-MaheUAPV #graph #kernel
- Extensions of marginalized graph kernels (PM, NU, TA, JLP, JPV).
- ICML-2004-MannorMHK #abstraction #clustering #learning
- Dynamic abstraction in reinforcement learning via clustering (SM, IM, AH, UK).
- ICML-2004-MannorSST #bias #estimation
- Bias and variance in value function estimation (SM, DS, PS, JNT).
- ICML-2004-MarlinZ #collaboration #multi
- The multiple multiplicative factor model for collaborative filtering (BMM, RSZ).
- ICML-2004-MelvilleM #learning
- Diverse ensembles for active learning (PM, RJM).
- ICML-2004-MerkeS #approximate #convergence #learning #linear
- Convergence of synchronous reinforcement learning with linear function approximation (AM, RS).
- ICML-2004-MoralesS #behaviour #learning
- Learning to fly by combining reinforcement learning with behavioural cloning (EFM, CS).
- ICML-2004-NatteeSNO #first-order #learning #mining #multi
- Learning first-order rules from data with multiple parts: applications on mining chemical compound data (CN, SS, MN, TO).
- ICML-2004-NguyenS #clustering #learning #using
- Active learning using pre-clustering (HTN, AWMS).
- ICML-2004-NguyenWJ #classification #detection #distributed #kernel #using
- Decentralized detection and classification using kernel methods (XN, MJW, MIJ).
- ICML-2004-OngMCS #kernel #learning
- Learning with non-positive kernels (CSO, XM, SC, AJS).
- ICML-2004-PeltonenSK #finite
- Sequential information bottleneck for finite data (JP, JS, SK).
- ICML-2004-PhillipsDS #approach #modelling
- A maximum entropy approach to species distribution modeling (SJP, MD, RES).
- ICML-2004-PieterN #learning
- Apprenticeship learning via inverse reinforcement learning (PA, AYN).
- ICML-2004-Potts #incremental #learning #linear
- Incremental learning of linear model trees (DP).
- ICML-2004-QiMPG #automation #predict
- Predictive automatic relevance determination by expectation propagation (Y(Q, TPM, RWP, ZG).
- ICML-2004-RayP #algorithm
- Sequential skewing: an improved skewing algorithm (SR, DP).
- ICML-2004-RosalesAF #clustering #learning #using
- Learning to cluster using local neighborhood structure (RR, KA, BJF).
- ICML-2004-RosencrantzGT #learning #predict
- Learning low dimensional predictive representations (MR, GJG, ST).
- ICML-2004-Rosset
- Model selection via the AUC (SR).
- ICML-2004-RuckertK #bound #learning #towards
- Towards tight bounds for rule learning (UR, SK).
- ICML-2004-RudarySP #adaptation #constraints #learning #reasoning
- Adaptive cognitive orthotics: combining reinforcement learning and constraint-based temporal reasoning (MRR, SPS, MEP).
- ICML-2004-Ryabko #learning #online
- Online learning of conditionally I.I.D. data (DR).
- ICML-2004-ScullyML
- Coalition calculation in a dynamic agent environment (TS, MGM, GL).
- ICML-2004-Shalev-ShwartzSN #learning #online #pseudo
- Online and batch learning of pseudo-metrics (SSS, YS, AYN).
- ICML-2004-SimsekB #abstraction #identification #learning #using
- Using relative novelty to identify useful temporal abstractions in reinforcement learning (ÖS, AGB).
- ICML-2004-SminchisescuJ #embedded #generative #modelling #visual notation
- Generative modeling for continuous non-linearly embedded visual inference (CS, ADJ).
- ICML-2004-Strens #performance #policy
- Efficient hierarchical MCMC for policy search (MJAS).
- ICML-2004-SuD #automation #component #probability
- Automated hierarchical mixtures of probabilistic principal component analyzers (TS, JGD).
- ICML-2004-SuttonRM #modelling #probability #random #sequence
- Dynamic conditional random fields: factorized probabilistic models for labeling and segmenting sequence data (CAS, KR, AM).
- ICML-2004-SzepesvariS
- Interpolation-based Q-learning (CS, WDS).
- ICML-2004-TaoSVO #approximate #learning #multi
- SVM-based generalized multiple-instance learning via approximate box counting (QT, SDS, NVV, TTO).
- ICML-2004-TaskarCK #learning #markov #network
- Learning associative Markov networks (BT, VC, DK).
- ICML-2004-ToutanovaMN #dependence #learning #modelling #random #word
- Learning random walk models for inducing word dependency distributions (KT, CDM, AYN).
- ICML-2004-TsochantaridisHJA #machine learning
- Support vector machine learning for interdependent and structured output spaces (IT, TH, TJ, YA).
- ICML-2004-VuralD #multi
- A hierarchical method for multi-class support vector machines (VV, JGD).
- ICML-2004-WeinbergerSS #kernel #learning #matrix #reduction
- Learning a kernel matrix for nonlinear dimensionality reduction (KQW, FS, LKS).
- ICML-2004-WellingRT #approximate #markov
- Approximate inference by Markov chains on union spaces (MW, MRZ, YWT).
- ICML-2004-WierstraW #markov #modelling
- Utile distinction hidden Markov models (DW, MW).
- ICML-2004-WingateS #named #parallel
- P3VI: a partitioned, prioritized, parallel value iterator (DW, KDS).
- ICML-2004-WuD #data flow
- Improving SVM accuracy by training on auxiliary data sources (PW, TGD).
- ICML-2004-XingSJ #process #type inference
- Bayesian haplo-type inference via the dirichlet process (EPX, RS, MIJ).
- ICML-2004-Ye #approximate #matrix #rank
- Generalized low rank approximations of matrices (JY).
- ICML-2004-YeJLP #analysis #feature model #linear
- Feature extraction via generalized uncorrelated linear discriminant analysis (JY, RJ, QL, HP).
- ICML-2004-Zadrozny #bias #classification #learning
- Learning and evaluating classifiers under sample selection bias (BZ).
- ICML-2004-Zhang #algorithm #linear #predict #probability #problem #scalability #using
- Solving large scale linear prediction problems using stochastic gradient descent algorithms (TZ0).
- ICML-2004-ZhangKY #algorithm
- Surrogate maximization/minimization algorithms for AdaBoost and the logistic regression model (ZZ, JTK, DYY).
- ICML-2004-ZhangY #estimation #probability
- Probabilistic score estimation with piecewise logistic regression (JZ, YY).
- ICML-2004-ZhangYK #algorithm #kernel #learning #matrix #using
- Bayesian inference for transductive learning of kernel matrix using the Tanner-Wong data augmentation algorithm (ZZ, DYY, JTK).

40 ×#learning

13 ×#classification

13 ×#multi

12 ×#kernel

11 ×#clustering

10 ×#algorithm

10 ×#using

7 ×#modelling

6 ×#feature model

6 ×#linear

13 ×#classification

13 ×#multi

12 ×#kernel

11 ×#clustering

10 ×#algorithm

10 ×#using

7 ×#modelling

6 ×#feature model

6 ×#linear