Relational Gaussian Processes for Learning Preference Relations
نویسنده
چکیده
Preference learning has received increasing attention in both machine learning and information retrieval. The goal of preference learning is to automatically learn a model to rank entities (e.g., documents, webpages, products, music, etc.) according to their degrees of relevance. The particularity of preference learning might be that the training data is a set of pairwise preferences between entities, instead of explicit entity-wise values. For example, we may only know that a user prefers an item to another one ei ≻ ej , but we do not know the exact preference degrees of items.
منابع مشابه
Learning Preferences with Hidden Common Cause Relations
Gaussian processes have successfully been used to learn preferences among entities as they provide nonparametric Bayesian approaches for model selection and probabilistic inference. For many entities encountered in real-world applications, however, there are complex relations between them. In this paper, we present a preference model which incorporates information on relations among entities. S...
متن کاملMulti-Relational Learning with Gaussian Processes
Due to their flexible nonparametric nature, Gaussian process models are very effective at solving hard machine learning problems. While existing Gaussian process models focus on modeling one single relation, we present a generalized GP model, named multi-relational Gaussian process model, that is able to deal with an arbitrary number of relations in a domain of interest. The proposed model is a...
متن کاملMining Preference Relations to Rank Complex Objects
One of the key tasks in data mining and information retrieval is to learn preference relations between objects. Approaches reported in the literature mainly aim at learning preference relations between objects represented according to the classical attribute-value representation. However, the growing interest in data mining techniques able to directly mine data represented according to more sop...
متن کاملFast Active Exploration for Link-Based Preference Learning Using Gaussian Processes
In preference learning, the algorithm observes pairwise relative judgments (preference) between items as training data for learning an ordering of all items. This is an important learning problem for applications where absolute feedback is difficult to elicit, but pairwise judgments are readily available (e.g., via implicit feedback [13]). While it was already shown that active learning can eff...
متن کاملGaussian Process Models for Colored Graphs
Many real-world domains can naturally be represented as complex graphs, i.e., in terms of entities (nodes) and relations (edges) among them. In domains with multiple relations, represented as colored graphs, we may improve the quality of a model by exploiting the correlations among relations of different types. To this end, we develop a multi-relational Gaussian process (MRGP) model. The MRGP m...
متن کامل