Gaussian Processes for Machine Learning (GPML) Toolbox

By Carl E. Rasmussen, Hannes Nickisch. 2010

1,098 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Statistical and Nonlinear Physics

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

1. Bernhard Schölkopf
2. Sergey Levine
3. Yoshua Bengio
4. Eric P. Xing
5. Ruslan Salakhutdinov
6. Pieter Abbeel
7. Jan Peters
8. Andrew Gordon Wilson

In Computational Theory and Mathematics

1. Bernhard Schölkopf
2. Jan Peters
3. Kun Zhang
4. Dominik Janzing
5. Paris Perdikaris
6. Andrew Gordon Wilson
7. G. Em Karniadakis
8. Zoubin Ghahramani

In Statistical and Nonlinear Physics

1. G. Em Karniadakis
2. Paris Perdikaris
3. Guofei Pang
4. Bernhard Schölkopf
5. Eric P. Xing
6. Andrew Gordon Wilson
7. Zhiting Hu
8. Maizar Raissi

Citations in the corpus

• Neural-net-induced Gaussian process regression for function approximation and PDE solution (2018, neutral): “6) [14]) is employed for the optimization.”
• Distinguishing Cause from Effect Using Observational Data: Methods and Benchmarks (2014, methodology): “5 (2014-12-08) (Rasmussen and Nickisch, 2010) for GP regression and ITE v0.”
• Deep Kernel Learning (2015, methodology): “We implemented DNNs based on Caffe (Jia et al., 2014), a general deep learning platform, and KISS-GP (Wilson and Nickisch, 2015; Wilson et al., 2015) leveraging GPML (Rasmussen and Nickisch, 2010) 1.”
• Scalable Variational Gaussian Process Classification (2014, methodology): “For comparison we used the EP FITC implementation from the GPML toolbox [Rasmussen and Nickisch, 2010] which is generally considered to be amongst the best implementations of this algorithm.”
• Kernel Interpolation for Scalable Structured Gaussian Processes (KISS-GP) (2015, methodology): “Kronecker and Toeplitz methods exploit the existing structure of the GP covariance matrix K to scale up inference and learning without approximations.”
• The future of employment: How susceptible are jobs to computerisation? (2017, methodology): “We tested three Gaussian process classifiers using the GPML toolbox (Rasmussen and Nickisch, 2010) on our data, built around exponentiated quadratic, rational quadratic and linear covariances.”
• Thoughts on Massively Scalable Gaussian Processes (2015, methodology): “• Code will be available as part of the GPML package (Rasmussen and Nickisch, 2010), with demonstrations at http://www.cs.cmu.edu/ ~andrewgw/pattern .”
• Groundwater level prediction in arid areas using wavelet analysis and Gaussian process regression (2021, background): “GPR is able to model complex relationships between inputs and outputs and cope with a broad range of simulation behaviors of alternative models (Karami et al., 2018; Liu &West, 2009; O’Hagan, 2013; Rasmussen & Nickisch, 2010).”
• A Scalable Hierarchical Gaussian Process Classifier (2019, methodology): “The generalized FITC [23] and the full GP classifier with EP used in our experiments are the MATLAB implementations from the GPML library [63].”
• State Space Gaussian Processes with Non-Gaussian Likelihood (2018, methodology): “We are interested in models following Rasmussen & Nickisch (2010) that – starting from the Gaussian prior f = N(f |m,K) given by the GP – admit an approximate posterior of the form Q(f”