Sequential Cooperative Bayesian Inference

Authors: Junqi Wang, Pei Wang, Patrick Shafto

ICML 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We approach consistency, convergence, and stability using a combination of new analytical and empirical methods. Section 4 presents empirical results analyzing the sample efficiency of SCBI versus BI, showing convergence of SCBI is considerably faster. Section 5 presents the empirical results testing robustness of SCBI to perturbations. In this section, we present some empirical results comparing the sample efficiency of SCBI and BI. Our simulation is based on Monte Carlo method of 10^4 teaching sequences (for each single point plotted).
Researcher Affiliation Academia 1Co Da S Lab, Department of Math & CS, Rutgers University at Newark, New Jersey, USA. Correspondence to: Junqi Wang <junqi.wang@rutgers.edu>.
Pseudocode Yes Algorithm 1 SCBI, without assumption (iii )
Open Source Code No The paper does not contain any explicit statements or links indicating that source code for the described methodology is publicly available.
Open Datasets No The paper describes a "Grid World" model as an application example and performs simulations. It does not mention using or providing access to a publicly available dataset for training.
Dataset Splits No The paper describes simulation experiments for theoretical models (e.g., matrix analysis, Grid World). It does not mention specific training, validation, or test dataset splits for an empirical dataset.
Hardware Specification No The paper does not explicitly describe the hardware (e.g., specific CPU/GPU models, memory) used to run its experiments or simulations.
Software Dependencies No The paper does not provide specific version numbers for any software dependencies (e.g., libraries, frameworks, programming languages) used in the experiments.
Experiment Setup Yes We simulated for n = 2, 3, . . . , 50 with a size-10^10 Monte Carlo method for each n to calculate P and E. We sampled 10^8 different M of shape 10 m for each 2 m 10. With sampling 500 matrices independently, we simulate a teacher teaches 2000 times to round 30 for each matrix. Our simulation is based on Monte Carlo method of 10^4 teaching sequences (for each single point plotted).