Learning Relational Kalman Filtering

Authors: Jaesik Choi, Eyal Amir, Tianfang Xu, Albert Valocchi

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental In the experiments, we use multiple synthetic data sets and a real-world groundwater flow MODFLOW model, the Republican River Compact Association (RRCA) model (Mc Kusick 2003) as shown in Figure 1. The dataset extracted from the RRCA model includes a set of monthly measured heads (water level) at over 3,000 wells for 850 months. Not all wells are measured every month.
Researcher Affiliation Academia Jaesik Choi School of Electrical and Computer Engineering Ulsan National Institute of Science and Technology Ulsan, Korea Eyal Amir Department of Computer Science University of Illinois at Urbana-Champaign Urbana, IL, USA Tianfang Xu and Albert J. Valocchi Department of Civil and Environmental Engineering University of Illinois at Urbana-Champaign Urbana, IL, USA
Pseudocode Yes Algorithm 1 Learning RKF; Algorithm 2 LRKF-Regroup (Prediction w/ testing data); Algorithm 3 Merge Atom
Open Source Code Yes Source code is available at http://pail.unist.ac.kr/LRKF.
Open Datasets Yes In the experiments, we use multiple synthetic data sets and a real-world groundwater flow MODFLOW model, the Republican River Compact Association (RRCA) model (Mc Kusick 2003) as shown in Figure 1.
Dataset Splits No The paper describes how data was split into training and testing sets ('10% of all measurements (about 10,038) are reserved for testing. All other measurements are used for training.'), but it does not mention a separate validation set or specific details for one.
Hardware Specification No The paper does not provide specific details about the hardware used for running the experiments (e.g., CPU, GPU models, memory).
Software Dependencies No The paper does not provide specific software dependencies or version numbers (e.g., libraries, frameworks, programming language versions) used for the experiments.
Experiment Setup Yes In both experiments, we set k in Algorithm Merge Atom to be 4. That is, two state variables will be merged if they have the same observation numbers and types when at least one relational observation is made.