Semantic Kernel Forests from Multiple Taxonomies
نویسندگان
چکیده
When learning features for complex visual recognition problems, labeled image exemplars alone can be insufficient. While an object taxonomy specifying the categories’ semantic relationships could bolster the learning process, not all relationships are relevant to a given visual classification task, nor does a single taxonomy capture all ties that are relevant. In light of these issues, we propose a discriminative feature learning approach that leverages multiple hierarchical taxonomies representing different semantic views of the object categories (e.g., for animal classes, one taxonomy could reflect their phylogenic ties, while another could reflect their habitats). For each taxonomy, we first learn a tree of semantic kernels, where each node has a Mahalanobis kernel optimized to distinguish between the classes in its children nodes. Then, using the resulting semantic kernel forest, we learn class-specific kernel combinations to select only those relationships relevant to recognize each object class. To learn the weights, we introduce a novel hierarchical regularization term that further exploits the taxonomies’ structure. We demonstrate our method on challenging object recognition datasets, and show that interleaving multiple taxonomic views yields significant accuracy improvements.
منابع مشابه
On the Role of Derivational Processes in the Formation of Non-Taxonomic Classes of Lexical Units in Russian
The paper is focused on classes of lexical units which arise as a result of derivational processes – word formation and semantic transfers, acting either in isolation or together, on the basis of common semantic foundations that bind targets and sources of derivation. The lexical items which constitute the classes under study vary in their denotative characteristics and due to their categ...
متن کاملSemantic Interoperability in XBRL Reports via Collaborative Ontology Mappings and Linked Data Principles
In an increasingly digitalized world, financial institutions have become data hoarders and find it difficult to efficiently exploit and interpret data from various sources. XBRL provides a global standard to exchange financial information. Although XBRL solves the syntactic heterogeneity problem, it doesnt solve the semantic heterogeneity problem when involving multiple taxonomies. In this pape...
متن کاملSemantic Integration Approach to Efficient Business Data Supply Chain: Integration Approach to Interoperable XBRL
As an open standard for electronic communication of business and financial data, XBRL has the potential of improving the efficiency of the business data supply chain. A number of jurisdictions have developed different XBRL taxonomies as their data standards. Semantic heterogeneity exists in these taxonomies, the corresponding instances, and the internal systems that store the original data. Con...
متن کاملRapid Induction of Multiple Taxonomies for Enhanced Faceted Text Browsing
In this paper we present and compare two methodologies for rapidly inducing multiple subject-specific taxonomies from crawled data. The first method involves a sentence-level words co-occurrence frequency method for building the taxonomy, while the second involves the bootstrapping of a Word2Vec based algorithm with a directed crawler. We exploit the multilingual open-content directory of the W...
متن کاملLearning with Multiple Similarities Learning with Multiple Similarities
Title of dissertation: LEARNINGWITHMULTIPLE SIMILARITIES Abhishek Kumar, Doctor of Philosophy, 2013 Dissertation directed by: Professor Hal Daumé III Department of Computer Science The notion of similarities between data points is central to many classification and clustering algorithms. We often encounter situations when there are more than one set of pairwise similarity graphs between objects...
متن کامل