revrandhttp://mloss.orgUpdates and additions to revrandenSun, 29 Jan 2017 04:33:54 -0000revrand 1.0.0<html><p>This library implements various Bayesian linear models (Bayesian linear regression) and generalized linear models. A few features of this library are: </p> <ul> <li><p>A fancy basis functions/feature composition framework for combining basis functions like radial basis function, sigmoidal basis functions, polynomial basis functions etc. </p> </li> <li><p>Basis functions that can be used to approximate Gaussian processes with shift invariant covariance functions (e.g. square exponential) when used with linear models [1], [2], [3]. </p> </li> <li><p>Non-Gaussian likelihoods with Bayesian generalized linear models (GLMs). We infer all of the parameters in the GLMs using stochastic variational inference [4], and we approximate the posterior over the weights with a mixture of Gaussians, like [5]. </p> </li> <li><p>Large scale learning using stochastic gradient descent (Adam, AdaDelta and more). </p> </li> <li><p>Scikit Learn compatibility, i.e. usable with pipelines. </p> </li> <li><p>A host of decorators for scipy.optimize.minimize and stochastic gradients that enhance the functionality of these optimisers. </p> </li> </ul> <p>[1] Yang, Z., Smola, A. J., Song, L., &amp; Wilson, A. G. "A la Carte -- Learning Fast Kernels". Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, pp. 1098-1106, 2015. </p> <p>[2] Le, Q., Sarlos, T., &amp; Smola, A. "Fastfood-approximating kernel expansions in loglinear time." Proceedings of the international conference on machine learning. 2013. </p> <p>[3] Rahimi, A., &amp; Recht, B. "Random features for large-scale kernel machines." Advances in neural information processing systems. 2007. </p> <p>[4] Kingma, D. P., &amp; Welling, M. "Auto-encoding variational Bayes". Proceedings of the 2nd International Conference on Learning Representations (ICLR). 2014. </p> <p>[5] Gershman, S., Hoffman, M., &amp; Blei, D. "Nonparametric variational inference". Proceedings of the international conference on machine learning. 2012. </p></html>daniel steinberg, louis tiao, alistair reid, lachlan mccalman, david cole, simon ocallaghanSun, 29 Jan 2017 04:33:54 -0000 gradient descentlarge scale learningnonparametric bayesnonlinear regressiongaussian processesgeneralized linear modelssparkfast foodrandom features