IEEE Trans Image Process - Decomposition-based transfer distance metric learning for image classification.

Tópicos

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

Distance metric learning (DML) is a critical factor for image analysis and pattern recognition. To learn a robust distance metric for a target task, we need abundant side information (i.e., the similarity/dissimilarity pairwise constraints over the labeled data), which is usually unavailable in practice due to the high labeling cost. This paper considers the transfer learning setting by exploiting the large quantity of side information from certain related, but different source tasks to help with target metric learning (with only a little side information). The state-of-the-art metric learning algorithms usually fail in this setting because the data distributions of the source task and target task are often quite different. We address this problem by assuming that the target distance metric lies in the space spanned by the eigenvectors of the source metrics (or other randomly generated bases). The target metric is represented as a combination of the base metrics, which are computed using the decomposed components of the source metrics (or simply a set of random bases); we call the proposed method, decomposition-based transfer DML (DTDML). In particular, DTDML learns a sparse combination of the base metrics to construct the target metric by forcing the target metric to be close to an integration of the source metrics. The main advantage of the proposed method compared with existing transfer metric learning approaches is that we directly learn the base metric coefficients instead of the target metric. To this end, far fewer variables need to be learned. We therefore obtain more reliable solutions given the limited side information and the optimization tends to be faster. Experiments on the popular handwritten image (digit, letter) classification and challenge natural image annotation tasks demonstrate the effectiveness of the proposed method.

Resumo Limpo

distanc metric learn dml critic factor imag analysi pattern recognit learn robust distanc metric target task need abund side inform ie similaritydissimilar pairwis constraint label data usual unavail practic due high label cost paper consid transfer learn set exploit larg quantiti side inform certain relat differ sourc task help target metric learn littl side inform stateoftheart metric learn algorithm usual fail set data distribut sourc task target task often quit differ address problem assum target distanc metric lie space span eigenvector sourc metric random generat base target metric repres combin base metric comput use decompos compon sourc metric simpli set random base call propos method decompositionbas transfer dml dtdml particular dtdml learn spars combin base metric construct target metric forc target metric close integr sourc metric main advantag propos method compar exist transfer metric learn approach direct learn base metric coeffici instead target metric end far fewer variabl need learn therefor obtain reliabl solut given limit side inform optim tend faster experi popular handwritten imag digit letter classif challeng natur imag annot task demonstr effect propos method

Resumos Similares

IEEE Trans Pattern Anal Mach Intell - Online Multiple Kernel Similarity Learning for Visual Search. ( 0,746467543421625 )
J Biomed Inform - Active learning strategies for the deduplication of electronic patient data using classification trees. ( 0,736794834999993 )
J Chem Inf Model - SCISSORS: practical considerations. ( 0,724984548223487 )
Neural Comput - Exploitation of pairwise class distances for ordinal classification. ( 0,719725342799402 )
Neural Comput - Refined rademacher chaos complexity bounds with applications to the multikernel learning problem. ( 0,716182890839134 )
IEEE Trans Neural Netw Learn Syst - A Kernel Classification Framework for Metric Learning. ( 0,709477900545936 )
IEEE J Biomed Health Inform - Identifying mammalian MicroRNA targets based on supervised distance metric learning. ( 0,704852678471119 )
IEEE Trans Image Process - Self-supervised online metric learning with low rank constraint for scene categorization. ( 0,690957508784963 )
Comput. Biol. Med. - Robust prediction of protein subcellular localization combining PCA and WSVMs. ( 0,687730305405377 )
Neural Comput - Improved similarity measures for small sets of spike trains. ( 0,687228675053959 )
IEEE Trans Image Process - Artistic image analysis using graph-based learning approaches. ( 0,684736788017461 )
IEEE Trans Pattern Anal Mach Intell - Distance-Based Image Classification: Generalizing to New Classes at Near Zero Cost. ( 0,675002616237151 )
IEEE Trans Image Process - 3-D object retrieval and recognition with hypergraph analysis. ( 0,673253525129936 )
Int J Neural Syst - Aggregation of sparse linear discriminant analyses for event-related potential classification in brain-computer interface. ( 0,672999627407705 )
IEEE Trans Image Process - Person re-identification over camera networks using multi-task distance metric learning. ( 0,657841689256587 )
IEEE Trans Image Process - Multiple-kernel, multiple-instance similarity features for efficient visual object detection. ( 0,647989722762 )
Artif Intell Med - An evaluation of heuristics for rule ranking. ( 0,647140897657814 )
IEEE Trans Pattern Anal Mach Intell - Learning with Box Kernels. ( 0,641053562696111 )
J Biomed Inform - Incremental Gaussian Discriminant Analysis based on Graybill and Deal weighted combination of estimators for brain tumour diagnosis. ( 0,639002836301253 )
IEEE Trans Image Process - View-based discriminative probabilistic modeling for 3D object retrieval and recognition. ( 0,636094645556941 )
IEEE Trans Image Process - Fast semantic diffusion for large-scale context-based image and video annotation. ( 0,629284751087601 )
IEEE Trans Image Process - Fuzzy C-means clustering with local information and kernel metric for image segmentation. ( 0,629004328595077 )
IEEE Trans Image Process - Saliency and gist features for target detection in satellite images. ( 0,627194751762611 )
J Biomed Inform - Class proximity measures--dissimilarity-based classification and display of high-dimensional data. ( 0,621252973222876 )
IEEE Trans Image Process - Implicit polynomial representation through a fast fitting error estimation. ( 0,619014283810428 )
IEEE Trans Image Process - Label transfer by measuring compactness. ( 0,618867830941086 )
Comput. Biol. Med. - Similarity measure for quality control of dental CAD/CAM-applications. ( 0,613854390420162 )
IEEE Trans Image Process - Co-transduction for shape retrieval. ( 0,610586628890519 )
IEEE Trans Neural Netw Learn Syst - An efficient topological distance-based tree kernel. ( 0,609408873573566 )
Neural Comput - Multiple spectral kernel learning and a gaussian complexity computation. ( 0,607275898621641 )
AMIA Annu Symp Proc - Shortest Path Edit Distance for Enhancing UMLS Integration and Audit. ( 0,606926509850895 )
IEEE Trans Pattern Anal Mach Intell - Weakly Supervised Recognition of Daily Life Activities with Wearable Sensors. ( 0,60483103931546 )
Neural Comput - Computing sparse representations of multidimensional signals using Kronecker bases. ( 0,602983160979476 )
Neural Comput - EEG data space adaptation to reduce intersession nonstationarity in brain-computer interface. ( 0,602486355127006 )
IEEE Trans Image Process - Two-dimensional approach to full-reference image quality assessment based on positional structural information. ( 0,599224497794176 )
Comput Math Methods Med - On multilabel classification methods of incompletely labeled biomedical text data. ( 0,598686622252243 )
IEEE Trans Image Process - Hyperspectral image classification through bilayer graph-based learning. ( 0,598095109535366 )
Comput Math Methods Med - Developing image processing meta-algorithms with data mining of multiple metrics. ( 0,597582125897237 )
IEEE Trans Image Process - Data-dependent hashing based on p-stable distribution. ( 0,596721310421102 )
Comput Math Methods Med - Pulse waveform classification using support vector machine with Gaussian time warp edit distance kernel. ( 0,59601150903728 )
Neural Comput - Quantifying neurotransmission reliability through metrics-based information analysis. ( 0,59390435695392 )
IEEE Trans Image Process - Geodesic propagation for semantic labeling. ( 0,593267244626338 )
IEEE Trans Neural Netw Learn Syst - On recursive edit distance kernels with application to time series classification. ( 0,591993753218054 )
IEEE Trans Neural Netw Learn Syst - Ordinal Distance Metric Learning for Image Ranking. ( 0,591922739860843 )
J. Comput. Biol. - Imbalanced class learning in epigenetics. ( 0,590511769681649 )
IEEE Trans Image Process - Random forest construction with robust semisupervised node splitting. ( 0,589494516995182 )
J Chem Inf Model - ReactionPredictor: prediction of complex chemical reactions at the mechanistic level using machine learning. ( 0,586479117462103 )
IEEE Trans Neural Netw Learn Syst - Partially shared latent factor learning with multiview data. ( 0,586311380954969 )
J Chem Inf Model - Assessing relative bioactivity of chemical substances using quantitative molecular network topology analysis. ( 0,585925091434886 )
IEEE Trans Image Process - Two-direction nonlocal model for image denoising. ( 0,585708155738669 )
Neural Comput - Adaptive metric learning vector quantization for ordinal classification. ( 0,585671712949477 )
IEEE Trans Image Process - Linear time distances between fuzzy sets with applications to pattern matching and classification. ( 0,584075676412333 )
IEEE Trans Image Process - Multiple kernel sparse representations for supervised and unsupervised learning. ( 0,584021298237626 )
J Chem Inf Model - Atom environment kernels on molecules. ( 0,583164617976707 )
Neural Comput - Information-theoretic semi-supervised metric learning via entropy regularization. ( 0,582552754152156 )
Int J Neural Syst - Structurally enhanced incremental neural learning for image classification with subgraph extraction. ( 0,580551295629415 )
IEEE Trans Image Process - Incremental training of a detector using online sparse eigendecomposition. ( 0,579903911079153 )
Methods Inf Med - Probability machines: consistent probability estimation using nonparametric learning machines. ( 0,574027307639204 )
J Biomed Inform - Multi-label classification of chronically ill patients with bag of words and supervised dimensionality reduction algorithms. ( 0,571301479437074 )
IEEE Trans Neural Netw Learn Syst - ML-Tree: a tree-structure-based approach to multilabel learning. ( 0,570967051093809 )
J Biomed Inform - Portable automatic text classification for adverse drug reaction detection via multi-corpus training. ( 0,56903033814667 )
IEEE Trans Image Process - Manifold regularized multitask learning for semi-supervised multilabel image classification. ( 0,567970748368504 )
J Chem Inf Model - Learning to predict chemical reactions. ( 0,567510670726341 )
J Chem Inf Model - Note on naive Bayes based on binary descriptors in cheminformatics. ( 0,56738315816163 )
IEEE Trans Image Process - Task-specific image partitioning. ( 0,566932424962269 )
IEEE Trans Image Process - Multiview Hessian regularization for image annotation. ( 0,563909889275698 )
IEEE Trans Pattern Anal Mach Intell - Feature Selection with Conjunctions of Decision Stumps and Learning from Microarray Data. ( 0,563434486259052 )
IEEE Trans Pattern Anal Mach Intell - Representation Learning: A Review and New Perspectives. ( 0,562361590015202 )
IEEE Trans Pattern Anal Mach Intell - Learning Categories from Few Examples with Multi Model Knowledge Transfer. ( 0,561683921994997 )
IEEE Trans Image Process - Structured max-margin learning for inter-related classifier training and multilabel image annotation. ( 0,559171364837182 )
IEEE Trans Image Process - Image quality assessment using multi-method fusion. ( 0,556750795704898 )
Neural Comput - Strictly positive-definite spike train kernels for point-process divergences. ( 0,55663208641543 )
IEEE Trans Image Process - Active learning for solving the incomplete data problem in facial age classification by the furthest nearest-neighbor criterion. ( 0,556353237028145 )
J Chem Inf Model - Anatomy of high-performance 2D similarity calculations. ( 0,556005752990317 )
IEEE Trans Image Process - Joint segmentation of images and scanned point cloud in large-scale street scenes with low-annotation cost. ( 0,555978391949776 )
J Am Med Inform Assoc - Active learning for clinical text classification: is it better than random sampling? ( 0,555188488746036 )
Neural Comput - Online learning with (multiple) kernels: a review. ( 0,554943117459189 )
Neural Comput - Incremental learning by message passing in hierarchical temporal memory. ( 0,551507615154661 )
IEEE Trans Image Process - Image denoising with dominant sets by a coalitional game approach. ( 0,551123318826902 )
IEEE Trans Image Process - Real-time probabilistic covariance tracking with efficient model update. ( 0,550278168391236 )
J. Comput. Biol. - Phylogeny inference based on spectral graph clustering. ( 0,549212583770789 )
IEEE Trans Neural Netw Learn Syst - A Distributed Approach Toward Discriminative Distance Metric Learning. ( 0,548662124021924 )
Neural Comput - Adaptive multiclass classification for brain computer interfaces. ( 0,547371915294206 )
Int J Comput Assist Radiol Surg - Automatic scoring of virtual mastoidectomies using expert examples. ( 0,546371270255141 )
Int J Neural Syst - Span: spike pattern association neuron for learning spatio-temporal spike patterns. ( 0,545592280065996 )
J Chem Inf Model - QSPR prediction of the stability constants of gadolinium(III) complexes for magnetic resonance imaging. ( 0,545040186839665 )
Int J Comput Assist Radiol Surg - Three-dimensional skeletonization and symbolic description in vascular imaging: preliminary results. ( 0,544698331321953 )
Neural Comput - Multilabel classification with principal label space transformation. ( 0,543539910743718 )
J Biomed Inform - Comparing and combining chunkers of biomedical text. ( 0,543109246958538 )
Neural Comput - Neural decoding with kernel-based metric learning. ( 0,542774849356349 )
Neural Comput - Reduction from cost-sensitive ordinal ranking to weighted binary classification. ( 0,542762592882888 )
Comput Methods Programs Biomed - Multistage approach for clustering and classification of ECG data. ( 0,542522934842213 )
J Chem Inf Model - Fusing dual-event data sets for Mycobacterium tuberculosis machine learning models and their evaluation. ( 0,540527340058259 )
Brief. Bioinformatics - Similarity-based machine learning methods for predicting drug-target interactions: a brief review. ( 0,539196797345316 )
J Biomed Inform - Learning classification models from multiple experts. ( 0,538176910107784 )
J Med Syst - 3D similarity-dissimilarity plot for high dimensional data visualization in the context of biomedical pattern classification. ( 0,536879665969523 )
Comput Math Methods Med - Correlation kernels for support vector machines classification with applications in cancer data. ( 0,536171560363817 )
Comput Biol Chem - Ranking of microRNA target prediction scores by Pareto front analysis. ( 0,535194064373007 )
IEEE Trans Image Process - Robust image deblurring with an inaccurate blur kernel. ( 0,535014570427244 )
IEEE Trans Image Process - Semisupervised multiview distance metric learning for cartoon synthesis. ( 0,534837206256918 )