Collaborative Gaussian Processes for Preference Learning

By Neil Houlsby, José Miguel Hernández-Lobato, Ferenc Huszár, Zoubin Ghahramani. 2012

132 citations recorded by the source corpus.

Subfields: Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Probabilistic & Bayesian Methods

Citations in the corpus

Evidential Deep Learning to Quantify Classification Uncertainty (2018, background): “The value of their uncertainty calculation has set the state of the art in active learning [12].”
Probabilistic Backpropagation for Scalable Learning of Bayesian Neural Networks (2015, methodology): “As future work we plan to evaluate PBP in multi-label and multi-class problems and experiment with networks that include more than one hidden layer.”
Predictive Entropy Search for Efficient Global Optimization of Black-box Functions (2014, neutral): “This simpliﬁes the negative-deﬁnite constraint.”
Decomposition of Uncertainty in Bayesian Deep Learning for Efficient and Risk-sensitive Learning (2017, background): “Perhaps closest to our work is BALD (Houlsby et al., 2012), however because GPs are used, this approach cannot model problems with complex noise.”
Asymmetric Transfer Learning with Deep Gaussian Processes (2015, methodology): “This probabilistic nature and the predictive variance has also been used to develop simple, effective, and theoretically well-grounded active learning models (Houlsby et al., 2012).”
Ordered Preference Elicitation Strategies for Supporting Multi-Objective Decision Making (2018, background): “[25] for pairwise feedback from multiple users or Guo et al.”
Cold-start Active Learning with Robust Ordinal Matrix Factorization (2014, methodology): “Previously, it has been used for preference elicitation and is called Bayesian Active Learning by Disagree-ment (BALD) (Houlsby et al., 2012).”
Probabilistic machine learning and artificial intelligence (2015, methodology): “We call this the personalization of models, and it is naturally implemented using hierarchical Bayesian approaches such as hierarchical Dirichlet processes 36 , and Bayesian multitask learning 88,89 .”
Predictive Entropy Search for Bayesian Optimization with Unknown Constraints (2015, methodology): “Instead, we follow Houlsby et al. (2012); Hern ´ andez-Lobato et al. (2014) and take advantage of the symmetry of mutual information, rewriting this acquisition function as the mutual information between y and x (cid:63) given the collected data D N .”
Parallel Predictive Entropy Search for Batch Global Optimization of Expensive Objective Functions (2015, background): “, S t , x ∗ (cid:1) is the joint posterior predictive distibution for { y q } =1 given the observed data, D and the location of the global maximizer of f .”