Using Galaxy Evolution as Source of Physics-Based Ground Truth for Generative Models
- URL: http://arxiv.org/abs/2407.07229v1
- Date: Tue, 9 Jul 2024 21:01:08 GMT
- Title: Using Galaxy Evolution as Source of Physics-Based Ground Truth for Generative Models
- Authors: Yun Qi Li, Tuan Do, Evan Jones, Bernie Boscoe, Kevin Alfaro, Zooey Nguyen,
- Abstract summary: We build a conditional denoising diffusionaxy probabilistic model (DDPM) and a conditional variational autoencoder (CVAE)
This is one of the first studies to probe these generative models using physically motivated metrics.
We find that both models produce comparable realistic galaxies based on human evaluation, but our physics-based metrics are better able to discern the strengths and weaknesses of the generative models.
- Score: 0.9701233658865522
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Generative models producing images have enormous potential to advance discoveries across scientific fields and require metrics capable of quantifying the high dimensional output. We propose that astrophysics data, such as galaxy images, can test generative models with additional physics-motivated ground truths in addition to human judgment. For example, galaxies in the Universe form and change over billions of years, following physical laws and relationships that are both easy to characterize and difficult to encode in generative models. We build a conditional denoising diffusion probabilistic model (DDPM) and a conditional variational autoencoder (CVAE) and test their ability to generate realistic galaxies conditioned on their redshifts (galaxy ages). This is one of the first studies to probe these generative models using physically motivated metrics. We find that both models produce comparable realistic galaxies based on human evaluation, but our physics-based metrics are better able to discern the strengths and weaknesses of the generative models. Overall, the DDPM model performs better than the CVAE on the majority of the physics-based metrics. Ultimately, if we can show that generative models can learn the physics of galaxy evolution, they have the potential to unlock new astrophysical discoveries.
Related papers
- Physics-Inspired Generative Models in Medical Imaging: A Review [5.499670421465509]
Physics-inspired generative models, in particular diffusion and Poisson flow models, enhance Bayesian methods and promise great utilities in medical imaging.
Major applications of physics-inspired generative models in medical imaging are presented, comprising image reconstruction, image generation, and image analysis.
Future research directions are brainstormed, including unification of physics-inspired generative models, integration with vision-language models (VLMs), and potential novel applications of generative models.
arXiv Detail & Related papers (2024-07-15T16:08:22Z) - VideoPhy: Evaluating Physical Commonsense for Video Generation [93.28748850301949]
We present VideoPhy, a benchmark designed to assess whether the generated videos follow physical commonsense for real-world activities.
Specifically, we curate a list of 688 captions that involve interactions between various material types in the physical world.
We generate videos conditioned on these captions from diverse state-of-the-art text-to-video generative models, including open models (e.g., VideoCrafter2) and closed models (e.g., Lumiere from Google, Pika)
arXiv Detail & Related papers (2024-06-05T17:53:55Z) - Can AI Understand Our Universe? Test of Fine-Tuning GPT by Astrophysical Data [6.0108108767559525]
ChatGPT is the most talked-about concept in recent months, captivating both professionals and the general public alike.
In this article, we fine-tune the generative pre-trained transformer (GPT) model by the astronomical data from the observations of galaxies, quasars, stars, gamma-ray bursts (GRBs) and simulations of black holes (BHs)
We regard this as a successful test, marking the LLM's proven efficacy in scientific research.
arXiv Detail & Related papers (2024-04-14T20:52:19Z) - Towards solving model bias in cosmic shear forward modeling [2.967246997200238]
Weak gravitational lensing generates a slight shearing of galaxy morphologies called cosmic shear.
Modern techniques of shear estimation based on statistics of ellipticity measurements suffer from the fact that the ellipticity is not a well-defined quantity for arbitrary galaxy light profiles.
We show that a hybrid physical and deep learning Hierarchical Bayesian Model, where a generative model captures the galaxy morphology, enables us to recover an unbiased estimate of the shear on realistic galaxies.
arXiv Detail & Related papers (2022-10-28T16:23:49Z) - What Language Model to Train if You Have One Million GPU Hours? [54.32062236748831]
We study the impact of different modeling practices and their impact on zero-shot generalization.
We also study the performance of a multilingual model and how it compares to the English-only one.
All our models and code are open-sourced at https://huggingface.co/bigscience.
arXiv Detail & Related papers (2022-10-27T13:43:27Z) - Your Autoregressive Generative Model Can be Better If You Treat It as an
Energy-Based One [83.5162421521224]
We propose a unique method termed E-ARM for training autoregressive generative models.
E-ARM takes advantage of a well-designed energy-based learning objective.
We show that E-ARM can be trained efficiently and is capable of alleviating the exposure bias problem.
arXiv Detail & Related papers (2022-06-26T10:58:41Z) - Which priors matter? Benchmarking models for learning latent dynamics [70.88999063639146]
Several methods have proposed to integrate priors from classical mechanics into machine learning models.
We take a sober look at the current capabilities of these models.
We find that the use of continuous and time-reversible dynamics benefits models of all classes.
arXiv Detail & Related papers (2021-11-09T23:48:21Z) - Realistic galaxy image simulation via score-based generative models [0.0]
We show that a score-based generative model can be used to produce realistic yet fake images that mimic observations of galaxies.
Subjectively, the generated galaxies are highly realistic when compared with samples from the real dataset.
arXiv Detail & Related papers (2021-11-02T16:27:08Z) - Machine Learning for Discovering Effective Interaction Kernels between
Celestial Bodies from Ephemerides [10.77689830299308]
We use a data-driven learning approach to derive a stable and accurate model for the motion of celestial bodies in our Solar System.
By modeling the major astronomical bodies in the Solar System as pairwise interacting agents, our learned model generate extremely accurate dynamics.
Our model can provide a unified explanation to the observation data, especially in terms of reproducing the perihelion precession of Mars, Mercury, and the Moon.
arXiv Detail & Related papers (2021-08-26T16:30:59Z) - Physics-Integrated Variational Autoencoders for Robust and Interpretable
Generative Modeling [86.9726984929758]
We focus on the integration of incomplete physics models into deep generative models.
We propose a VAE architecture in which a part of the latent space is grounded by physics.
We demonstrate generative performance improvements over a set of synthetic and real-world datasets.
arXiv Detail & Related papers (2021-02-25T20:28:52Z) - Visual Grounding of Learned Physical Models [66.04898704928517]
Humans intuitively recognize objects' physical properties and predict their motion, even when the objects are engaged in complicated interactions.
We present a neural model that simultaneously reasons about physics and makes future predictions based on visual and dynamics priors.
Experiments show that our model can infer the physical properties within a few observations, which allows the model to quickly adapt to unseen scenarios and make accurate predictions into the future.
arXiv Detail & Related papers (2020-04-28T17:06:38Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.