Official pytorch implementation of the paper "Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels" (NeurIPS 2020)

Unofficial Implementation of the paper "Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Control", applied to gym environments

Hyperpatameter Bayesian Optimization for Image Classification in PyTorch

Bayesian Optimization for MPPI Control of Robot Arm Planar Pushing

We have created a module to run the Gaussian process model. We have implemented the code based on GPyTorch.

Dataset and code for "Coarse-Grained Density Functional Theory Predictions via Deep Kernel Learning"

A Fast and Simplified Python Library for Uncertainty Estimation

Uncertainty in convolutional neural network predictions using Gaussian processes

Topics include function approximation, learning dynamics, using learned dynamics in control and planning, handling uncertainty in learned models, learning from demonstration, and model-based and model-free reinforcement learning.

Proof-of-principle application of Gaussian process modeling to gamma-ray analyses. Code repository associated with the paper https://arxiv.org/abs/2010.10450.

Demonstrating the use of Prefect to orchestrate the creation of machine learning surrogate models as applied to mechanistic crop models.

Distributed surrogate-assisted evolutionary methods for multi-objective optimization of high-dimensional dynamical systems

Highly performant and scalable out-of-the-box gaussian process regression and Bernoulli classification. Built upon GPyTorch, with a familiar sklearn api.

Time series forecasting on an hourly energy dataset, with LSTM & Transformer models implemented in PyTorch Lightning. Deployment of the Transformer model using Docker, with GPU support.

Models for EthicML

Explores the application of Gaussian Process (GP) and sparse GP algorithms to handle missing heart rate time series dataset. Our findings emphasize the importance of kernel selection, specifically the RBF kernel, and the careful tuning of hyperparameters to achieve optimal performance in imputation tasks

Implementation of Gaussian Process (GP) models using GPyTorch.

Testing deployment of a Ray cluster on AWS

Evaluating Deep Gaussian processes

Spatiotemporal Gaussian process modeling for environmental data: non-stationary PDE prior, deep kernels, multi-fidelity fusion, and A-optimal sampling.非稳态 PDE + 核深度学习 + 多保真 Co-Kriging + 主动采样的物理约束克里金方法，用于复杂时空环境建模与预测