论文信息 - TensorBNN: Bayesian Inference for Neural Networks using Tensorflow

TensorBNN: Bayesian Inference for Neural Networks using Tensorflow

TensorBNN is a new package based on TensorFlow that implements Bayesian inference for modern neural network models. The posterior density of neural network model parameters is represented as a point cloud sampled using Hamiltonian Monte Carlo. The TensorBNN package leverages TensorFlow's architecture and training features as well as its ability to use modern graphics processing units (GPU) in both the training and prediction stages.

Michelle Kuchera | Harrison Prosper | Braden Kronheim

[1] Jian Sun,et al. Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[2] J. T. Childers,et al. Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC , 2012 .

[3] Eric T. Nalisnick. On Priors for Bayesian Neural Networks , 2018 .

[4] Geoffrey E. Hinton,et al. Bayesian Learning for Neural Networks , 1995 .

[5] Sanjiv Kumar,et al. On the Convergence of Adam and Beyond , 2018 .

[6] Daniel Foreman-Mackey,et al. emcee: The MCMC Hammer , 2012, 1202.3665.

[7] Eric Jones,et al. SciPy: Open Source Scientific Tools for Python , 2001 .

[8] The Cms Collaboration. Observation of a new boson at a mass of 125 GeV with the CMS experiment at the LHC , 2012, 1207.7235.

[9] Natalia Gimelshein,et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[10] Nando de Freitas,et al. Adaptive Hamiltonian and Riemann manifold Monte Carlo samplers , 2013, ICML 2013.