On the Identifiability of Hybrid Deep Generative Models: Meta-Learning as a Solution

Authors: Yubo Ye, Maryam Tolou, Sumeet Vadhavkar, Xiajun Jiang, Huafeng Liu, Linwei Wang

NeurIPS 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental On synthetic and real-data benchmarks, we provide strong empirical evidence for the un-identifiability of existing hybrid-DGMs using unconditional priors, and strong identifiability results of the presented meta-formulations of hybrid-DGMs.
Researcher Affiliation Academia Yubo Ye Zhejiang University 22230131@zju.edu.cn Maryam Toloubidokhti Rochester Institute of Technology mt6129@rit.edu Sumeet Vadhavkar Rochester Institute of Technology sv6234@rit.edu Xiajun Jiang Rochester Institute of Technology xj7056@rit.edu Huafeng Liu B Zhejiang University liuhf@zju.edu.cn Linwei Wang Rochester Institute of Technology linwei.wang@rit.edu
Pseudocode No The paper does not contain any pseudocode or explicitly labeled algorithm blocks.
Open Source Code No The data and code will be released soon.
Open Datasets Yes We considered three simulated and one real-world benchmarks for hybrid-DGMs, including three simulated physics systems of forced damped pendulum [2], advection-diffusion system [2] and double pendulum [4], and one real-world system double pendulum [4]. We used the dataset of a double pendulum introduced by [4].
Dataset Splits Yes We generated total 60000 samples and separated them into a training, validation, and test sets with 4,0000, 10000, and 1,0000 samples, respectively. We generated 20000, 5000, and 5000 training, validation, and test samples, respectively.
Hardware Specification No The paper does not provide specific hardware details (e.g., GPU/CPU models, memory, or cloud instances) used for running its experiments.
Software Dependencies No The paper does not provide specific software dependencies with version numbers.
Experiment Setup Yes For each of the three simulated physics systems, we randomly sampled the initial states and parameters of the governing function to generate observations. We divided each video into observations of x s to consist of 20 temporal frames with a sampling frequency of 100 Hz. Because there is no clear indication of which samples belong to the same data-generation process, we use 7 samples preceding the current query sample as the context samples. We adopted Equation (19) as the physics-component of the hybrid-DGM, using the known lengths of the two arms (L1 = 91mm and L2 = 70mm) and assuming m = 1.