# Matt Taddy

#### 7 packages on CRAN

This is a bare-bones implementation of sampling algorithms for a variety of Bayesian stick-breaking (marginally DP) mixture models, including particle learning and Gibbs sampling for static DP mixtures, particle learning for dynamic BAR stick-breaking, and DP mixture regression. The software is designed to be easy to customize to suit different situations and for experimentation with stick-breaking models. Since particles are repeatedly copied, it is not an especially efficient implementation.

Fast distributed/parallel estimation for multinomial logistic regression via Poisson factorization and the 'gamlr' package. For details see: Taddy (2015, AoAS), Distributed Multinomial Regression, <arXiv:1311.6139>.

The gamma lasso algorithm provides regularization paths corresponding to a range of non-convex cost functions between L0 and L1 norms. As much as possible, usage for this package is analogous to that for the glmnet package (which does the same thing for penalization between L1 and L2 norms). For details see: Taddy (2017 JCGS), 'One-Step Estimator Paths for Concave Regularization', <arXiv:1308.5623>.

Maximum a posteriori (MAP) estimation for topic models (i.e., Latent Dirichlet Allocation) in text analysis, as described in Taddy (2012) 'On estimation and selection for topic models'. Previous versions of this code were included as part of the 'textir' package. If you want to take advantage of openmp parallelization, uncomment the relevant flags in src/MAKEVARS before compiling.

Multinomial (inverse) regression inference for text documents and associated attributes. For details see: Taddy (2013 JASA) Multinomial Inverse Regression for Text Analysis <arXiv:1012.2098> and Taddy (2015, AoAS), Distributed Multinomial Regression, <arXiv:1311.6139>. A minimalist partial least squares routine is also included. Note that the topic modeling capability of earlier 'textir' is now a separate package, 'maptpx'.

Inference by sequential Monte Carlo for dynamic tree regression and classification models with hooks provided for sequential design and optimization, fully online learning with drift, variable selection, and sensitivity analysis of inputs. Illustrative examples from the original dynamic trees paper are facilitated by demos in the package; see demo(package="dynaTree").

Bayesian nonstationary, semiparametric nonlinear regression and design by treed Gaussian processes (GPs) with jumps to the limiting linear model (LLM). Special cases also implemented include Bayesian linear models, CART, treed linear models, stationary separable and isotropic GPs, and GP single-index models. Provides 1-d and 2-d plotting functions (with projection and slice capabilities) and tree drawing, designed for visualization of tgp-class output. Sensitivity analysis and multi-resolution models are supported. Sequential experimental design and adaptive sampling functions are also provided, including ALM, ALC, and expected improvement. The latter supports derivative-free optimization of noisy black-box functions.