Active Learning with Gaussian Processes for Object Categorization

By Ashish Kapoor, Kristen Grauman, Raquel Urtasun, Trevor Darrell. 2007

413 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Non-Linear Regression for Bag-of-Words Data via Gaussian Process Latent Variable Set Model (2015, methodology): “Thus, the idea of the GP-LVSM can be applied to various machine learning problems, which have been solved based on GP regression such as multi-task learning (Bonilla, Chai, and Williams 2008) and active learning (Kapoor and Grauman 2007).”
Reliable Uncertainty Estimates in Deep Neural Networks using Noise Contrastive Priors (2018, methodology): “Kapoor et al. (2007) estimate expected improvement using a Gaussian process.”
Multi-Level Active Prediction of Useful Image Annotations for Recognition (2008, background): “Recent w ork has considered active learning with Gaussian Process classifiers [20], and relevance feedback f or video annotations [21].”
The Power of Ensembles for Active Learning in Image Classification (2018, background): “Active learning concepts relevant for this paper include uncertainty sampling ([42], [32], [35]), query-by-committee ([30], [5]) and density-based approaches ([52], [61], [46]).”
Adversarial Active Learning for Deep Networks: a Margin Based Approach (2018, background): “Uncertainty selection may be also tailored to network ensemble, either by disagreement over the models ( Query by committee , (Seung et al., 1992)) or by sampling through the distribution of the weights ( Bayesian active learning , (Kapoor et al., 2007)).”
What Do I Annotate Next? An Empirical Study of Active Learning for Action Localization (2018, background): “As emphasized in previous work [37,46], the key challenge in active learning is to design the proper selection function, which seeks to minimize the number of times an oracle is queried to reach a target performance.”
Approximations for Binary Gaussian Process Classification (2008, methodology): “…prediction (Chu et al., 2005), EP for affect recognition (Kapoor and Picard, 2005), VB for weld cracking prognosis (Gibbs and MacKay, 2000), Label Regression (LR) serves for object categorization (Kapoor et al., 2007) and MCMC sampling is applied to rheuma diagnosis (Schwaighofer et al., 2002).”
Active Learning for Convolutional Neural Networks: A Core-Set Approach (2017, methodology): “Bayesian active learning methods typically use a non-parametric model like Gaussian process to estimate the expected improvement by each query (Kapoor et al., 2007) or the expected error after a set of queries (Roy & McCallum, 2001).”
Active learning of visual descriptors for grasping using non-parametric smoothed beta distributions (2012, neutral): “[42,43].”
Entropy-based active learning for object recognition (2008, methodology): “A recent paper [3] uses active learning and Pyramid Match Kernels to improve object category recognition.”
Multi-fidelity classification using Gaussian processes: accelerating the prediction of large-scale computational models (2019, methodology): “Since Gaussian process classiﬁers have an associated variance with their predictions, they naturally enable the design of e ﬀ ective data acquisition policies via active learning [15, 6].”
Gaussian Processes for Object Categorization (2010, background): “1This paper expands on our previous conference publication (Kapoor et al. 2007); here we provide further explanation of our Gaussian Process model, extend it to allow combinations of multiple kernel functions, and report and discuss a number of additional experiments.”
Cost-Effective Active Learning for Deep Image Classification (2017, background): “works [7], [24] is proposed to take the samples to maximize the increase of mutual information between the candidate instance and the remaining unlabeled instances under the Gaussian process framework.”
Learning by Asking Questions (2017, background): “Active learning (AL) involves a collection of unlabeled examples and a learner that selects which samples will be labeled by an oracle [26, 33, 48, 53].”