Modeling Continuous Stochastic Processes with Dynamic Normalizing Flows

Authors: Ruizhi Deng, Bo Chang, Marcus A. Brubaker, Greg Mori, Andreas Lehrmann

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

Reproducibility Variable Result LLM Response
Research Type Experimental In this section, we apply our models on synthetic data generated from common continuous-time stochastic processes and complex real-world datasets. The proposed CTFP and latent CTFP models are compared against two baseline models: latent ODEs [42] and variational RNNs (VRNNs) [12].
Researcher Affiliation Collaboration Ruizhi Deng1,2 Bo Chang1 Marcus A. Brubaker1,3,4 Greg Mori1,2 Andreas M. Lehrmann1 1Borealis AI 2Simon Fraser University 3York University 4Vector Institute
Pseudocode No The paper presents mathematical formulations and descriptions of its model but does not include any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes Please see the supplementary materials for details about our experimental setup and model implementations.
Open Datasets Yes Mujoco-Hopper [42] consists of 10,000 sequences that are simulated by a Hopper model from the Deep Mind Control Suite in a Mu Jo Co environment [46]. PTB Diagnostic Database (PTBDB) [4] consists of excerpts of ambulatory electrocardiography (ECG) recordings. Beijing Air-Quality Dataset (BAQD) [47] is a dataset consisting of multi-year recordings of weather and air quality data across different locations in Beijing.
Dataset Splits No The paper mentions 'training' and 'test' sets with specific parameters (e.g., 'λtrain = 2' and 'λtest = 20'), but it does not explicitly provide details about a separate 'validation' split or its size/methodology in the main text.
Hardware Specification No The paper does not provide specific hardware details such as exact GPU/CPU models, processor types, or memory amounts used for running the experiments in the main text.
Software Dependencies No The paper mentions the use of GRU as the RNN cell but does not provide specific version numbers for any software dependencies or libraries.
Experiment Setup Yes For latent ODE, latent CTFP, and VRNN, we report the (upper bound of) NLL estimated by the IWAE bound [6] in Equation 13, using K = 25 samples of latent variables. For CTFP, the reported values are exact; for the other three models, we report IWAE bounds using K = 125 samples.