Particle Flow Bayes’ Rule

Authors: Xinshi Chen, Hanjun Dai, Le Song

ICML 2019 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental 4. Experiments We conduct experiments on multivariate Gaussian model, hidden Markov model and Bayesian logistic regression to demonstrate the generalization ability of PFBR and also its accuracy for posterior estimation.
Researcher Affiliation Collaboration 1School of Mathematics, 2School of Computational Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia, USA. 3Ant Financial, Hangzhou, China.
Pseudocode Yes Algorithm 1 Overall Learning Algorithm
Open Source Code No The paper does not contain any explicit statement about releasing source code for the described methodology or a link to a code repository.
Open Datasets Yes Bayesian Logistic Regression (BLR). We consider logistic regression for digits classification on the MNIST8M 8 vs. 6 dataset which contains about 1.6M training samples and 1932 testing samples.
Dataset Splits No The paper mentions 'Perform a validation step on Dvali validation' in Algorithm 1, but does not provide specific details on the size or methodology of the validation split for any of its experiments.
Hardware Specification No The paper mentions 'gpu' in Table 1 caption, but does not provide specific details such as exact GPU/CPU models, processor types, or memory amounts used for running experiments.
Software Dependencies No The paper does not provide specific software dependencies or library versions (e.g., Python 3.8, PyTorch 1.9) required to replicate the experiments.
Experiment Setup Yes Since we use a batch size of 128 and consider 10 stages, the first gradient step of our method starts after around 103 samples are visited. [...] In our experiment, we use µx = 0, Σx = I and Σo = 3I. [...] We use 256 obtained particles as samples from p(x|Om) and compare it with true posteriors.