IEEE Trans Image Process - Random forest construction with robust semisupervised node splitting.

Tópicos

{ learn(2355) train(1041) set(1003) }
{ featur(1941) imag(1645) propos(1176) }
{ data(3008) multipl(1320) sourc(1022) }
{ structur(1116) can(940) graph(676) }
{ error(1145) method(1030) estim(1020) }
{ problem(2511) optim(1539) algorithm(950) }
{ perform(999) metric(946) measur(919) }
{ sampl(1606) size(1419) use(1276) }
{ framework(1458) process(801) describ(734) }
{ compound(1573) activ(1297) structur(1058) }
{ imag(1947) propos(1133) code(1026) }
{ studi(2440) review(1878) systemat(933) }
{ data(1714) softwar(1251) tool(1186) }
{ spatial(1525) area(1432) region(1030) }
{ imag(2675) segment(2577) method(1081) }
{ surgeri(1148) surgic(1085) robot(1054) }
{ general(901) number(790) one(736) }
{ import(1318) role(1303) understand(862) }
{ model(2656) set(1616) predict(1553) }
{ data(2317) use(1299) case(1017) }
{ method(2212) result(1239) propos(1039) }
{ can(774) often(719) complex(702) }
{ measur(2081) correl(1212) valu(896) }
{ method(1219) similar(1157) match(930) }
{ featur(3375) classif(2383) classifi(1994) }
{ assess(1506) score(1403) qualiti(1306) }
{ treatment(1704) effect(941) patient(846) }
{ concept(1167) ontolog(924) domain(897) }
{ method(984) reconstruct(947) comput(926) }
{ search(2224) databas(1162) retriev(909) }
{ model(3480) simul(1196) paramet(876) }
{ intervent(3218) particip(2042) group(1664) }
{ activ(1138) subject(705) human(624) }
{ health(1844) social(1437) communiti(874) }
{ cancer(2502) breast(956) screen(824) }
{ drug(1928) target(777) effect(648) }
{ implement(1333) system(1263) develop(1122) }
{ method(1969) cluster(1462) data(1082) }
{ model(3404) distribut(989) bayesian(671) }
{ data(1737) use(1416) pattern(1282) }
{ inform(2794) health(2639) internet(1427) }
{ system(1976) rule(880) can(841) }
{ imag(1057) registr(996) error(939) }
{ bind(1733) structur(1185) ligand(1036) }
{ sequenc(1873) structur(1644) protein(1328) }
{ imag(2830) propos(1344) filter(1198) }
{ network(2748) neural(1063) input(814) }
{ patient(2315) diseas(1263) diabet(1191) }
{ take(945) account(800) differ(722) }
{ motion(1329) object(1292) video(1091) }
{ chang(1828) time(1643) increas(1301) }
{ clinic(1479) use(1117) guidelin(835) }
{ algorithm(1844) comput(1787) effici(935) }
{ extract(1171) text(1153) clinic(932) }
{ method(1557) propos(1049) approach(1037) }
{ design(1359) user(1324) use(1319) }
{ control(1307) perform(991) simul(935) }
{ model(2220) cell(1177) simul(1124) }
{ care(1570) inform(1187) nurs(1089) }
{ case(1353) use(1143) diagnosi(1136) }
{ howev(809) still(633) remain(590) }
{ data(3963) clinic(1234) research(1004) }
{ studi(1410) differ(1259) use(1210) }
{ risk(3053) factor(974) diseas(938) }
{ research(1085) discuss(1038) issu(1018) }
{ system(1050) medic(1026) inform(1018) }
{ model(2341) predict(2261) use(1141) }
{ visual(1396) interact(850) tool(830) }
{ perform(1367) use(1326) method(1137) }
{ studi(1119) effect(1106) posit(819) }
{ blood(1257) pressur(1144) flow(957) }
{ record(1888) medic(1808) patient(1693) }
{ health(3367) inform(1360) care(1135) }
{ monitor(1329) mobil(1314) devic(1160) }
{ ehr(2073) health(1662) electron(1139) }
{ state(1844) use(1261) util(961) }
{ research(1218) medic(880) student(794) }
{ patient(2837) hospit(1953) medic(668) }
{ age(1611) year(1155) adult(843) }
{ medic(1828) order(1363) alert(1069) }
{ signal(2180) analysi(812) frequenc(800) }
{ cost(1906) reduc(1198) effect(832) }
{ group(2977) signific(1463) compar(1072) }
{ gene(2352) biolog(1181) express(1162) }
{ first(2504) two(1366) second(1323) }
{ time(1939) patient(1703) rate(768) }
{ patient(1821) servic(1111) care(1106) }
{ use(2086) technolog(871) perceiv(783) }
{ can(981) present(881) function(850) }
{ analysi(2126) use(1163) compon(1037) }
{ high(1669) rate(1365) level(1280) }
{ use(976) code(926) identifi(902) }
{ use(1733) differ(960) four(931) }
{ result(1111) use(1088) new(759) }
{ survey(1388) particip(1329) question(1065) }
{ estim(2440) model(1874) function(577) }
{ decis(3086) make(1611) patient(1517) }
{ process(1125) use(805) approach(778) }
{ activ(1452) weight(1219) physic(1104) }
{ detect(2391) sensit(1101) algorithm(908) }

Resumo

Random forest (RF) is a very important classifier with applications in various machine learning tasks, but its promising performance heavily relies on the size of labeled training data. In this paper, we investigate constructing of RFs with a small size of labeled data and find that the performance bottleneck is located in the node splitting procedures; hence, existing solutions fail to properly partition the feature space if there are insufficient training data. To achieve robust node splitting with insufficient data, we present semisupervised splitting to overcome this limitation by splitting nodes with the guidance of both labeled and abundant unlabeled data. In particular, an accurate quality measure of node splitting is obtained by carrying out the kernel-based density estimation, whereby a multiclass version of asymptotic mean integrated squared error criterion is proposed to adaptively select the optimal bandwidth of the kernel. To avoid the curse of dimensionality, we project the data points from the original high-dimensional feature space onto a low-dimensional subspace before estimation. A unified optimization framework is proposed to select a coupled pair of subspace and separating hyperplane such that the smoothness of the subspace and the quality of the splitting are guaranteed simultaneously. Our algorithm efficiently avoids overfitting caused by bad initialization and local maxima when compared with conventional margin maximization-based semisupervised methods. We demonstrate the effectiveness of the proposed algorithm by comparing it with state-of-the-art supervised and semisupervised algorithms for typical computer vision applications, such as object categorization, face recognition, and image segmentation, on publicly available data sets.

Resumo Limpo

random forest rf import classifi applic various machin learn task promis perform heavili reli size label train data paper investig construct rfs small size label data find perform bottleneck locat node split procedur henc exist solut fail proper partit featur space insuffici train data achiev robust node split insuffici data present semisupervis split overcom limit split node guidanc label abund unlabel data particular accur qualiti measur node split obtain carri kernelbas densiti estim wherebi multiclass version asymptot mean integr squar error criterion propos adapt select optim bandwidth kernel avoid curs dimension project data point origin highdimension featur space onto lowdimension subspac estim unifi optim framework propos select coupl pair subspac separ hyperplan smooth subspac qualiti split guarante simultan algorithm effici avoid overfit caus bad initi local maxima compar convent margin maximizationbas semisupervis method demonstr effect propos algorithm compar stateoftheart supervis semisupervis algorithm typic comput vision applic object categor face recognit imag segment public avail data set

Resumos Similares

J Chem Inf Model - Note on naive Bayes based on binary descriptors in cheminformatics. ( 0,739744975318035 )
IEEE Trans Neural Netw Learn Syst - A Distributed Approach Toward Discriminative Distance Metric Learning. ( 0,716743280563242 )
IEEE Trans Image Process - Multiple kernel sparse representations for supervised and unsupervised learning. ( 0,716026753092315 )
IEEE Trans Image Process - Stable orthogonal local discriminant embedding for linear dimensionality reduction. ( 0,708779136750967 )
IEEE Trans Image Process - Structured max-margin learning for inter-related classifier training and multilabel image annotation. ( 0,701331284410513 )
IEEE Trans Neural Netw Learn Syst - Semi-supervised domain adaptation on manifolds. ( 0,697959358138117 )
IEEE Trans Image Process - Sparse representation with kernels. ( 0,686896468477608 )
Comput Math Methods Med - A novel multiinstance learning approach for liver cancer recognition on abdominal CT images based on CPSO-SVM and IO. ( 0,683553519156944 )
IEEE Trans Image Process - Task-specific image partitioning. ( 0,680526502853263 )
IEEE Trans Image Process - Object detection with DoG scale-space: a multiple kernel learning approach. ( 0,676601292923942 )
IEEE Trans Image Process - Visual classification with multitask joint sparse representation. ( 0,675739380284769 )
IEEE Trans Pattern Anal Mach Intell - Feature Selection with Conjunctions of Decision Stumps and Learning from Microarray Data. ( 0,673930711370273 )
IEEE Trans Image Process - Robust face recognition with structurally incoherent low-rank matrix decomposition. ( 0,665451576592783 )
IEEE Trans Pattern Anal Mach Intell - Feature Selection and Kernel Learning for Local Learning-Based Clustering. ( 0,661608719857299 )
IEEE Trans Image Process - Blur and illumination robust face recognition via set-theoretic characterization. ( 0,652601831280122 )
IEEE Trans Pattern Anal Mach Intell - Trainable Convolution Filters and Their Application to Face Recognition. ( 0,647610595345794 )
IEEE Trans Image Process - Image segmentation by MAP-ML estimations. ( 0,645311119715259 )
Int J Neural Syst - Aggregation of sparse linear discriminant analyses for event-related potential classification in brain-computer interface. ( 0,644560305261148 )
IEEE Trans Neural Netw Learn Syst - An efficient topological distance-based tree kernel. ( 0,642324002955959 )
IEEE Trans Image Process - Sparse demixing of hyperspectral images. ( 0,642211336334696 )
IEEE Trans Image Process - Semisupervised multiview distance metric learning for cartoon synthesis. ( 0,640855042696856 )
IEEE Trans Image Process - Discriminant learning through multiple principal angles for visual recognition. ( 0,63584445292237 )
IEEE Trans Neural Netw Learn Syst - Ordinal Distance Metric Learning for Image Ranking. ( 0,634804492884263 )
IEEE Trans Image Process - Multimodal graph-based reranking for web image search. ( 0,633580155497465 )
IEEE Trans Image Process - Artistic image analysis using graph-based learning approaches. ( 0,630904361470663 )
IEEE Trans Image Process - Real-time probabilistic covariance tracking with efficient model update. ( 0,62971089631288 )
IEEE Trans Pattern Anal Mach Intell - Learning Categories from Few Examples with Multi Model Knowledge Transfer. ( 0,629365109088756 )
IEEE Trans Image Process - Coupled kernel embedding for low resolution face image recognition. ( 0,629039388399171 )
IEEE Trans Image Process - Grassmannian regularized structured multi-view embedding for image classification. ( 0,626243127366761 )
IEEE Trans Image Process - Multiview Hessian regularization for image annotation. ( 0,624722695754637 )
J Biomed Inform - Class proximity measures--dissimilarity-based classification and display of high-dimensional data. ( 0,624517954695429 )
IEEE Trans Image Process - From local pixel structure to global image super-resolution: a new face hallucination framework. ( 0,621310644904676 )
IEEE Trans Pattern Anal Mach Intell - Learning with Augmented Features for Supervised and Semi-supervised Heterogeneous Domain Adaptation. ( 0,61949875597318 )
IEEE Trans Image Process - Nonnegative Tensor Co-Factorization and Its Unified Solution. ( 0,617952271803007 )
IEEE Trans Pattern Anal Mach Intell - Label Consistent K-SVD: Learning A Discriminative Dictionary for Recognition. ( 0,613363820577806 )
IEEE Trans Image Process - A linear support higher-order tensor machine for classification. ( 0,609524150793271 )
IEEE Trans Image Process - Incremental training of a detector using online sparse eigendecomposition. ( 0,609309643294131 )
IEEE Trans Pattern Anal Mach Intell - Constrained Nonnegative Matrix Factorization for Image Representation. ( 0,606414337668448 )
Neural Comput - Multiple spectral kernel learning and a gaussian complexity computation. ( 0,604692653408303 )
IEEE Trans Vis Comput Graph - A Structure-Based Distance Metric for High-Dimensional Space Exploration with Multi-Dimensional Scaling. ( 0,604056328512684 )
IEEE Trans Image Process - Geodesic propagation for semantic labeling. ( 0,603007796735667 )
IEEE Trans Image Process - Self-supervised online metric learning with low rank constraint for scene categorization. ( 0,602758682291748 )
Neural Comput - Adaptive multiclass classification for brain computer interfaces. ( 0,600649844509539 )
IEEE Trans Image Process - Cooperative sparse representation in two opposite directions for semi-supervised image annotation. ( 0,600032970973816 )
IEEE Trans Pattern Anal Mach Intell - Facial Age Estimation by Learning from Label Distributions. ( 0,594947483331578 )
Neural Comput - Large margin low rank tensor analysis. ( 0,594766147343508 )
IEEE Trans Image Process - Design and optimization of color lookup tables on a simplex topology. ( 0,593275262535764 )
Comput. Biol. Med. - Locally linear representation Fisher criterion based tumor gene expressive data classification. ( 0,59209844874604 )
IEEE Trans Image Process - CSMMI: Class-Specific Maximization of Mutual Information for Action and Gesture Recognition. ( 0,591444166046456 )
IEEE Trans Neural Netw Learn Syst - MTC: A Fast and Robust Graph-Based Transductive Learning Method. ( 0,590635816326773 )
IEEE Trans Image Process - Decomposition-based transfer distance metric learning for image classification. ( 0,589494516995182 )
IEEE Trans Image Process - Subspaces indexing model on Grassmann manifold for image search. ( 0,587854191833174 )
IEEE Trans Image Process - Ordinal feature selection for iris and palmprint recognition. ( 0,587717118114438 )
IEEE Trans Image Process - Manifold regularized multitask learning for semi-supervised multilabel image classification. ( 0,587139882935969 )
IEEE Trans Image Process - Harmonic active contours. ( 0,584339475113113 )
Neural Comput - Multilabel classification with principal label space transformation. ( 0,584090122078635 )
IEEE Trans Pattern Anal Mach Intell - Trinary-Projection Trees for Approximate Nearest Neighbor Search. ( 0,58250759391915 )
IEEE Trans Pattern Anal Mach Intell - Consistent Latent Position Estimation and Vertex Classification for Random Dot Product Graphs. ( 0,580802477236518 )
IEEE Trans Neural Netw Learn Syst - Online Sequential Extreme Learning Machine With Kernels. ( 0,580173313196375 )
Int J Comput Assist Radiol Surg - Statistical shape model of a liver for autopsy imaging. ( 0,580150122424826 )
IEEE Trans Neural Netw Learn Syst - A Kernel Classification Framework for Metric Learning. ( 0,578947991465125 )
IEEE Trans Image Process - Conjunctive patches subspace learning with side information for collaborative image retrieval. ( 0,578269333269425 )
IEEE Trans Neural Netw Learn Syst - ML-Tree: a tree-structure-based approach to multilabel learning. ( 0,577627997227426 )
IEEE Trans Image Process - Face identification using large feature sets. ( 0,577627633713581 )
Neural Comput - Robust subspace discovery via relaxed rank minimization. ( 0,575613054458494 )
Comput Math Methods Med - Dimensionality reduction by supervised neighbor embedding using laplacian search. ( 0,574950811915438 )
IEEE Trans Pattern Anal Mach Intell - Weakly Supervised Recognition of Daily Life Activities with Wearable Sensors. ( 0,574949459899715 )
IEEE Trans Neural Netw Learn Syst - Learning With Mixed Hard/Soft Pointwise Constraints. ( 0,57102613966791 )
Neural Comput - Incremental learning by message passing in hierarchical temporal memory. ( 0,570046938875592 )
IEEE Trans Image Process - An attribute-assisted reranking model for web image search. ( 0,569220348653088 )
IEEE Trans Image Process - Local learned dictionaries optimized to edge orientation for inverse halftoning. ( 0,567910945518297 )
IEEE Trans Image Process - Toward a unified color space for perception-based image processing. ( 0,563844084140351 )
Neural Comput - Computing sparse representations of multidimensional signals using Kronecker bases. ( 0,563458529728134 )
IEEE Trans Vis Comput Graph - SuperMatching: Feature Matching Using Supersymmetric Geometric Constraints. ( 0,562527633544604 )
IEEE Trans Image Process - Unified structured learning for simultaneous human pose estimation and garment attribute classification. ( 0,562186950989574 )
IEEE Trans Image Process - Two-dimensional maximum local variation based on image euclidean distance for face recognition. ( 0,56204489463479 )
Comput. Biol. Med. - Robust prediction of protein subcellular localization combining PCA and WSVMs. ( 0,561475546816951 )
IEEE Trans Image Process - Adaptive dictionary learning in sparse gradient domain for image recovery. ( 0,561429735531022 )
IEEE Trans Image Process - Local energy pattern for texture classification using self-adaptive quantization thresholds. ( 0,56090748865922 )
IEEE Trans Pattern Anal Mach Intell - Representation Learning: A Review and New Perspectives. ( 0,560045070529297 )
IEEE Trans Image Process - A unified framework for multioriented text detection and recognition. ( 0,559572291019635 )
IEEE Trans Image Process - Saliency and gist features for target detection in satellite images. ( 0,559366078511769 )
IEEE Trans Image Process - Unsupervised amplitude and texture classification of SAR images with multinomial latent model. ( 0,559161424552884 )
IEEE Trans Neural Netw Learn Syst - Partially shared latent factor learning with multiview data. ( 0,558715966708426 )
IEEE Trans Image Process - Active learning for solving the incomplete data problem in facial age classification by the furthest nearest-neighbor criterion. ( 0,558702159307781 )
IEEE Trans Image Process - Data-dependent hashing based on p-stable distribution. ( 0,558058139742418 )
Int J Neural Syst - Retrieval of noisy fingerprint patterns using metric attractor networks. ( 0,557514406469907 )
IEEE Trans Image Process - Learning discriminative dictionary for group sparse representation. ( 0,557061925482013 )
Neural Comput - Reduction from cost-sensitive ordinal ranking to weighted binary classification. ( 0,556056665715553 )
IEEE Trans Pattern Anal Mach Intell - A Bag-of-Features Framework to Classify Time Series. ( 0,555025971138899 )
IEEE Trans Image Process - Sparse tensor discriminant analysis. ( 0,554755694721646 )
IEEE Trans Image Process - Retina verification system based on biometric graph matching. ( 0,554522219126793 )
Artif Intell Med - Local binary patterns variants as texture descriptors for medical image analysis. ( 0,554466451241514 )
IEEE Trans Image Process - Color constancy using 3D scene geometry derived from a single image. ( 0,554070078743556 )
IEEE Trans Image Process - Image decomposition with multilabel context: algorithms and applications. ( 0,552513707060202 )
Neural Comput - Refined rademacher chaos complexity bounds with applications to the multikernel learning problem. ( 0,552397699514327 )
IEEE Trans Pattern Anal Mach Intell - LDAHash: Improved Matching with Smaller Descriptors. ( 0,552173360044752 )
IEEE Trans Image Process - Task-dependent visual-codebook compression. ( 0,551145285262019 )
IEEE Trans Image Process - Nonnegative local coordinate factorization for image representation. ( 0,550734000270302 )
IEEE Trans Neural Netw Learn Syst - On recursive edit distance kernels with application to time series classification. ( 0,549262756951636 )