Filters








5,741 Hits in 6.7 sec

Active learning and discovery of object categories in the presence of unnameable instances

Christoph Kading, Alexander Freytag, Erik Rodner, Paul Bodesheim, Joachim Denzler
2015 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
To meet these observations, we present a variant of the expected model output change principle for active learning and discovery in the presence of unnameable instances.  ...  In active learning, all categories occurring in collected data are usually assumed to be known in advance and experts should be able to label every requested instance.  ...  Active learning with unnameable instances Problem setting A very common assumption in active learning is that the oracle (e.g., a human annotator) can provide a label for every instance of the set of unlabeled  ... 
doi:10.1109/cvpr.2015.7299063 dblp:conf/cvpr/KadingFRBD15 fatcat:bfwbsrgksrfrdmv3amsnh74mhe

Large-Scale Active Learning with Approximations of Expected Model Output Changes [chapter]

Christoph Käding, Alexander Freytag, Erik Rodner, Andrea Perino, Joachim Denzler
2016 Lecture Notes in Computer Science  
At the same time, it outperforms previous active learning approaches in academic and real-world scenarios.  ...  Besides recent progress, it remains one of the fundamental challenges in computer vision and machine learning.  ...  However, rare class discovery is especially challenging in the presence of unnameable instances and not within the scope of this paper.  ... 
doi:10.1007/978-3-319-45886-1_15 fatcat:wo4hmfzhx5fztdznejeqftdili

Interactively building a discriminative vocabulary of nameable attributes

Devi Parikh, Kristen Grauman
2011 CVPR 2011  
The system takes object/scene-labeled images as input, and returns as output a set of attributes elicited from human annotators that distinguish the categories of interest.  ...  To ensure a compact vocabulary and efficient use of annotators' effort, we 1) show how to actively augment the vocabulary such that new attributes resolve inter-class confusions, and 2) propose a novel  ...  Acknowledgements: This research is supported in part by the Luce Foundation and NSF IIS-1065390.  ... 
doi:10.1109/cvpr.2011.5995451 dblp:conf/cvpr/ParikhG11 fatcat:nyxcokgayrbz7fhdihjqidhmtq

Discovering localized attributes for fine-grained recognition

Kun Duan, D. Parikh, D. Crandall, K. Grauman
2012 2012 IEEE Conference on Computer Vision and Pattern Recognition  
labels and object bounding boxes.  ...  "white belly"), but the question of how to choose these local attributes remains largely unexplored.  ...  Acknowledgments: The authors thank Dhruv Batra for discussions on the L-CRF formulation, and acknowledge support from NSF IIS-1065390, the Luce Foundation, the Lilly Endowment, and the IU Data-to-Insight  ... 
doi:10.1109/cvpr.2012.6248089 dblp:conf/cvpr/DuanPCG12 fatcat:ih5iacw3cvb5fgd2n2cnt7p2ma

A Semantic Reasoning Method Towards Ontological Model for Automated Learning Analysis [chapter]

Kingsley Okoye, Abdel-Rahman H. Tawil, Usman Naeem, Elyes Lamine
2015 Advances in Intelligent Systems and Computing  
the consistency of learning object/data types.  ...  The method applies semantic rules and description logic queries to build ontology model capable of automatically computing the various learning activities within a Learning Knowledge-Base, and to check  ...  Ny the number of instances for which learning activity y holds. And Nx˰̭ y is the number of instances for which activity x and y holds.  ... 
doi:10.1007/978-3-319-27400-3_5 fatcat:nitduqmuendotn4dccr7ymw3ni

Labels Direct Infants' Attention to Commonalities during Novel Category Learning

Nadja Althaus, Denis Mareschal, Andrew Bremner
2014 PLoS ONE  
Detailed analyses of infants' looking patterns during learning revealed that only infants who heard labels exhibited a rapid focus on the object part successive exemplars had in common.  ...  Both labeling and non-labeling phrases facilitated category formation in 12-month-olds but not 8month-olds (Experiment 1). Non-linguistic sounds did not produce this effect (Experiment 2).  ...  Acknowledgments We are grateful to Teodora Gliga for help with data collection, to Sarah Lloyd-Fox for recording auditory stimuli, and to the parents and infants who took part in this study.  ... 
doi:10.1371/journal.pone.0099670 pmid:25014254 pmcid:PMC4094422 fatcat:dupvlwv53nexbj4bdbblg2bu6q

Accommodating surprise in taxonomic tasks: the role of expertise

E Alberdi
2000 Cognitive Science  
Five professional botanists were asked to specify a category from a set of positive and negative instances.  ...  The target category in the study was defined by a feature that was unusual, hence situations of uncertainty and puzzlement were generated.  ...  Caroline Green helped in the rating of the protocols.  ... 
doi:10.1016/s0364-0213(99)00021-x fatcat:umjhru4aejh6rhvumupgn4mbxe

Accommodating Surprise in Taxonomic Tasks: The Role of Expertise

Eugenio Alberdi, Derek H. Sleeman, Meg Korpi
2000 Cognitive Science  
Five professional botanists were asked to specify a category from a set of positive and negative instances.  ...  The target category in the study was defined by a feature that was unusual, hence situations of uncertainty and puzzlement were generated.  ...  Caroline Green helped in the rating of the protocols.  ... 
doi:10.1207/s15516709cog2401_2 fatcat:6ois7vhgjfcczkc4p3253uejse

Deep Learning for Material recognition: most recent advances and open challenges [article]

Alain Tremeau, Sixiang Xu, Damien Muselet
2020 arXiv   pre-print
While deep neural networks provide very good results on object recognition and has been the topic of a huge amount of papers in the last decade, their adaptation to material images still requires some  ...  Nevertheless, recent studies achieve very good results in material recognition with deep learning and we propose, in this paper, to review most of them by focusing on three aspects: material image datasets  ...  The latest advancements in machine learning domain have highly revolutionized computational and data-minded methodologies used for materials design innovation and materials discovery and optimization  ... 
arXiv:2012.07495v1 fatcat:uujh5mrdzzhhhitc3fub3wiaye

AL-DDCNN: a distributed crossing semantic gap learning for person re-identification

Keyang Cheng, Yongzhao Zhan, Man Qi
2016 Concurrency and Computation  
Experiments show that the proposed approach achieve state-of-the-art recognition performance in the VIPeR data set and is with a good semantic explanation which cannot be given by other methods.  ...  To overcome the model's weakness in computing seep, parallelized implementations such as distributed parameter manipulation and attributes learning are employed in AL-DDCNN model.  ...  Acknowledgment This research is supported by the national science foundation of China (NFSC) No.61170126, the science foundation of Jiangsu province No.BK20150527 and the science foundation of Zhenjiang  ... 
doi:10.1002/cpe.3766 fatcat:zku34sdpunccbevnszt6wzb3z4

Unraveling the Taste Fabric of Social Networks

Hugo Liu, Pattie Maes, Glorianna Davenport
2006 International Journal on Semantic Web and Information Systems (IJSWIS)  
., and machine learning was applied to infer a semantic fabric of taste.  ...  For example, consider that "rock climbing," "yoga," the food "sushi," the music of "Mozart," and the books of "Ralph Waldo Emerson" all have something in common.  ...  Acknowledgement This research was supported by a British Telecom Fellowship, an AOL Fellowship, and by the research consortia sponsors of the MIT Media Lab.  ... 
doi:10.4018/jswis.2006010102 fatcat:jli7bqxfdnfsrijzp6cdt3dfhq

THE SIMILARITY-IN-TOPOGRAPHY PRINCIPLE: RECONCILING THEORIES OF CONCEPTUAL DEFICITS

W. Kyle Simmons, Lawrence W. Barsalou
2003 Cognitive Neuropsychology  
Barsalou. 1 "Category" will refer to a set of exemplars in the world, and "concept" will refer to a cognitive representation of the category in the brain.  ...  Depending on the level and location of a lesion in this system, a wide variety of deficits is possible.  ...  If the instances of a category share many correlated properties, its instances lump tightly in semantic space, whereas a category with diverse instances is distributed more broadly.  ... 
doi:10.1080/02643290342000032 pmid:20957580 fatcat:ij2ppnyoezftfku463ohjuruyi

Text analytics for life science using the Unstructured Information Management Architecture

R. Mack, S. Mukherjea, A. Soffer, N. Uramoto, E. Brown, A. Coden, J. Cooper, A. Inokuchi, B. Iyer, Y. Mass, H. Matsuzawa, L. V. Subramaniam
2004 IBM Systems Journal  
Acknowledgments BioTeKS is in large part a systems integration effort that builds on technologies and expertise developed  ...  Knowledge discovery can increase the speed (and hence the productivity) of a drug researcher finding a drug target, a competitor s patent activity, or a participant in a clinical trial.  ...  Some of the advantages of this tool are its interactive visual user interface for guiding the process of identifying true instances of an entity category based on confidence levels (estimates of in-class  ... 
doi:10.1147/sj.433.0490 fatcat:altfinouzbdy7mrcqx2kzdzecy

Introduction to Linked Data and Its Lifecycle on the Web [chapter]

Axel-Cyrille Ngonga Ngomo, Sören Auer, Jens Lehmann, Amrapali Zaveri
2014 Lecture Notes in Computer Science  
In this article we introduce the main concepts of Linked Data.  ...  With Linked Data, a very pragmatic approach towards achieving the vision of the Semantic Web has gained some traction in the last years.  ...  of OntoWiki as described in Section 4.  ... 
doi:10.1007/978-3-319-10587-1_1 fatcat:2vba6rydjvehro3cwcakafnn5e

A genomic data resource for predicting antimicrobial resistance from laboratory-derived antimicrobial susceptibility phenotypes

Margo VanOeffelen, Marcus Nguyen, Derya Aytan-Aktug, Thomas Brettin, Emily M Dietrich, Ronald W Kenyon, Dustin Machi, Chunhong Mao, Robert Olson, Gordon D Pusch, Maulik Shukla, Rick Stevens (+5 others)
2021 Briefings in Bioinformatics  
In addition to using the data to track the evolution and spread of AMR, it also serves as a useful starting point for building machine learning models for predicting AMR phenotypes.  ...  In this paper, we describe the characteristics of this collection, highlighting areas where sampling is comparatively deep or shallow, and showing areas where attention is needed from the research community  ...  We are extremely grateful to the members of the research community who are publishing AST data and genomes for use in the public domain.  ... 
doi:10.1093/bib/bbab313 pmid:34379107 pmcid:PMC8575023 fatcat:z2mxaakqs5earkcz3si33m5msq
« Previous Showing results 1 — 15 out of 5,741 results