GOFMM

Geometry Oblivious Fast Multipole method

GOFMM is an algorithm for compressing kernel matrices in arbitrary dimensions. GOFMM can also compress arbitrary symmetric positive definite matrices. GOFMM supports shared and distributed memory parallelism. To our knowledge GOFMM is the only open source software with such capabilities.

hIPPYlib

Inverse Problem PYthon library

hIPPYlib implements state-of-the-art scalable adjoint-based algorithms for PDE-based deterministic and Bayesian inverse problems. It builds on FEniCS for the discretization of the PDE and on PETSc for scalable and efficient linear algebra operations and solvers.

KNN-DBSCAN

KNN-DBSCAN is a shared and distributed memory parallel library for linkage-based point clustering in high-dimensions. Unlike the original DBSCAN which uses range search nearest neighbors, KNN-DBSCAN uses approximate k-nearest neighbors and results in better scalability.

Operator Inference model reduction

Python implementation of Operator Inference for model reduction

This is a Python implementation of Operator Inference for learning projection-based polynomial reduced-order models of dynamical systems. The procedure is data-driven and non-intrusive, making it a viable candidate for model reduction of "glass-box" systems. The package includes various tutorial examples.

MUQ

MIT Uncertainty Quantification Library

MUQ (MIT Uncertainty Quantification) is a C++/Python library for uncertainty quantification—in particular, for connecting complex models with UQ tools in a way that exposes model structure to the algorithms. MUQ is designed both for use by application scientists and engineers and as a platform for algorithm developers. It currently includes a wide variety of UQ capabilities: advanced Markov chain Monte Carlo algorithms for inference; approximation methods for computationally intensive likelihoods and forward models; adaptive methods (e.g., sparse polynomial approximations) for uncertainty propagation, global sensitivity analysis, and surrogate construction; and many others. MUQ optimizes UQ workflows through the use of directed acyclic graphs for dependency management. The underlying dependency graph enables structure-exploiting algorithms to cache and share information in a relatively transparent fashion. MUQ also operates seamlessly with packages such as FEniCS, libMesh, SUNDIALS, and NLopt.

TransportMaps

MIT TransportMaps library

TransportMaps is a Python (2.7/3.x) package for the construction of deterministic multi-dimensional couplings, induced by transport maps, between distributions. These couplings can be used for robust, fast, and accurate integration with respect to the complex distributions arising in Bayesian statistical models. They can also be used for density estimation and for sequential inference in state space models (e.g., data assimilation). The transport map method allows for an effective control over the accuracy of the inference, the parallel generation of Monte Carlo samples and quadrature rules, and the design of algorithms that make use of low-dimensional structure.

NR_SMOCU_SGD_GPC

This Python package provides the implementation of efficient Bayesian active learning algorithms for Gaussian Process classifiers (GPC). The acquisition function based on soft mean objective cost of uncertainty (SMOCU) can be estimated efficiently by the 1-d integral of the joint distribution of query label pairs without incrementally retraining GPC, which enables the first stochastic gradient descent (SGD) algorithm to reduce SMOCU for Bayesian query synthesis active learning algorithm.

Sync

Sync is a Python package demonstrating the optimal experimental design (OED) framework based on the mean objective cost of uncertainty (MOCU) for complex uncertain systems described by interconnected ordinary differential equations. MOCU-based OED is applied to the design of experiments that can effectively reduce the cost of achieving robust synchronization of an uncertain Kumamoto oscillator model.

SDE-MOCU

SDE-MOCU hosts a set of Matlab code that demonstrates our implementations of robust linear filtering and experimental design for physical systems governed by stochastic differential equations (SDEs) under model uncertainty. A sequential experimental design framework is built upon SDE-based physical models based on the mean objective cost of uncertainty derived based on the prior scientific knowledge.

Soft_MOCU

This Python package provides the implementation of efficient Bayesian active learning algorithms based on soft mean objective cost of uncertainty (Soft_MOCU). Soft_MOCU provides a strictly concave approximation of the original MOCU-based acquisition function for pool-based active learning, alleviating the myopic behavior of sequential Bayesian active learning for training optimal Bayesian classifiers (OBC).

WMOCU_AL

This Python package provides the implementation of efficient Bayesian active learning algorithms based on weighted mean objective cost of uncertainty (WMOCU). WMOCU is a strictly concave approximation of the original MOCU-based acquisition function for pool-based active learning, alleviating the myopic behavior of sequential Bayesian active learning for training optimal Bayesian classifiers (OBC).

BSDA

BSDA is a set of Python programs for the implementation of a generative optimal Bayesian supervised domain adaptation (BSDA) model that can integrate RNA sequencing (RNA-Seq) data from different domains along with their labels for improving prediction accuracy for subtyping in a given disease population.