Generative Neurosymbolic Machines
Authors: Jindong Jiang, Sungjin Ahn
NeurIPS 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | In experiments, we show that the proposed model significantly outperforms the previous structured representation models as well as the state-of-the-art non-structured generative models in terms of both structure accuracy and image generation quality. 4 Experiments Goals and Datasets. The goals of the experiments are (i) to evaluate the quality and properties of the generated images in terms of clarity and scene structure, (ii) to understand the factors of the datasets and hyperparameters that affect the performance, and (iii) to perform ablation studies to understand the key factors in the proposed architecture. |
| Researcher Affiliation | Academia | Jindong Jiang Department of Computer Science Rutgers University jindong.jiang@rutgers.edu Sungjin Ahn Department of Computer Science Rutgers University sjn.ahn@gmail.com |
| Pseudocode | No | The paper describes its models and methods in prose and mathematical formulations (e.g., Equation 1, 2, 3, 4) and provides a graphical model in Figure 1, but it does not include any explicitly labeled pseudocode or algorithm blocks. |
| Open Source Code | No | The paper does not provide any explicit statements about releasing open-source code or include a link to a code repository for the methodology described. |
| Open Datasets | Yes | We use the following three datasets: MNIST-4. In this dataset, an image is partitioned into four areas (top-right, top-left, bottom-right, and bottom-left), and one MNIST digit is placed in each quadrant. ... MNIST-10. To evaluate the effect of the number of components and complexity of the dependency structure, we also created a similar dataset containing ten MNIST-digits... Arrow Room. This dataset contains four 3D objects in a 3D space similar to CLEVR [25]. |
| Dataset Splits | No | The paper mentions training the model and evaluating performance but does not specify exact percentages or sample counts for training, validation, or test splits. It refers to 'training steps' and 'early training steps' and implicitly uses a validation process via a 'discriminability score' but without defining specific splits. |
| Hardware Specification | No | The paper does not provide any specific details about the hardware used for running the experiments, such as GPU models, CPU types, or memory specifications. |
| Software Dependencies | No | The paper does not specify any software dependencies or their version numbers, such as programming languages, libraries, or frameworks used for implementation. |
| Experiment Setup | Yes | We set the default drawing steps of GNM and Conv DRAW to 4 but also tested with 8 steps. ... As shown in Figure 4, the scene structure accuracy of Conv DRAW and VAE improved as the beta value increases. However, even for the largest value β = 10, the structure accuracy of Conv DRAW is still lower than (for ARROW room) or similar to (for MNIST-10) GNM with β = 1 while their log-likelihoods are significantly degraded. For GNM, we only tested β = [1, 2, 3] for MNIST-10 as it provides good and robust performance for these low values. ... We apply curriculum training to deal with the racing condition between the background and component modules, both trying to explain the full observation. For this, we suppress the learning of the background network in the early training steps and give a preference to the foreground modules to explain the scene. |