Insights on Galaxy Evolution from Interpretable Sparse Feature Networks

• URL: http://arxiv.org/abs/2501.00089v1
• Date: Mon, 30 Dec 2024 19:00:00 GMT
• Title: Insights on Galaxy Evolution from Interpretable Sparse Feature Networks
• Authors: John F. Wu,
• Abstract summary: We present a novel neural network architecture called a Sparse Feature Network (SFNet) SFNets produce interpretable features that can be linearly combined in order to estimate galaxy properties like optical emission line ratios or gas-phase metallicity. We find that SFNets do not sacrifice accuracy in order to gain interpretability, and that they perform comparably well to cutting-edge models on astronomical machine learning tasks.
• Score: 0.0
• License:
• Abstract: Galaxy appearances reveal the physics of how they formed and evolved. Machine learning models can now exploit galaxies' information-rich morphologies to predict physical properties directly from image cutouts. Learning the relationship between pixel-level features and galaxy properties is essential for building a physical understanding of galaxy evolution, but we are still unable to explicate the details of how deep neural networks represent image features. To address this lack of interpretability, we present a novel neural network architecture called a Sparse Feature Network (SFNet). SFNets produce interpretable features that can be linearly combined in order to estimate galaxy properties like optical emission line ratios or gas-phase metallicity. We find that SFNets do not sacrifice accuracy in order to gain interpretability, and that they perform comparably well to cutting-edge models on astronomical machine learning tasks. Our novel approach is valuable for finding physical patterns in large datasets and helping astronomers interpret machine learning results.

Related papers

• Learning the Evolution of Physical Structure of Galaxies via Diffusion Models [29.01313417459577]
  This paper introduces a novel approach to conditioning Denoising Diffusion Probabilistic Models (DDPM) on redshifts for generating galaxy images. We explore whether this advanced generative model can accurately capture the physical characteristics of galaxies based solely on their images and redshift measurements.
  arXiv  Detail & Related papers  (2024-11-27T15:28:30Z)
• Coding schemes in neural networks learning classification tasks [52.22978725954347]
  We investigate fully-connected, wide neural networks learning classification tasks. We show that the networks acquire strong, data-dependent features. Surprisingly, the nature of the internal representations depends crucially on the neuronal nonlinearity.
  arXiv  Detail & Related papers  (2024-06-24T14:50:05Z)
• Super Consistency of Neural Network Landscapes and Learning Rate Transfer [72.54450821671624]
  We study the landscape through the lens of the loss Hessian. We find that certain spectral properties under $mu$P are largely independent of the size of the network. We show that in the Neural Tangent Kernel (NTK) and other scaling regimes, the sharpness exhibits very different dynamics at different scales.
  arXiv  Detail & Related papers  (2024-02-27T12:28:01Z)
• Preliminary Report on Mantis Shrimp: a Multi-Survey Computer Vision Photometric Redshift Model [0.431625343223275]
  Photometric redshift estimation is a well-established subfield of astronomy. Mantis Shrimp is a computer vision model for photometric redshift estimation that fuses ultra-violet (GALEX), optical (PanSTARRS), and infrared (UnWISE) imagery.
  arXiv  Detail & Related papers  (2024-02-05T21:44:19Z)
• Galaxy Classification: A machine learning approach for classifying shapes using numerical data [0.0]
  We present a machine learning model for galaxy classification using numerical data from the Galaxy Zoo project. Our results show that our model achieves high accuracy in classifying galaxies and has the potential to significantly enhance our understanding of the formation and evolution of galaxies.
  arXiv  Detail & Related papers  (2023-11-30T20:47:16Z)
• Discovering Galaxy Features via Dataset Distillation [7.121183597915665]
  In many applications, Neural Nets (NNs) have classification performance on par or even exceeding human capacity. Here, we apply this idea to the notoriously difficult task of galaxy classification. We present a novel way to summarize and visualize prototypical galaxy morphology through the lens of neural networks.
  arXiv  Detail & Related papers  (2023-11-29T12:39:31Z)
• Seeing in Words: Learning to Classify through Language Bottlenecks [59.97827889540685]
  Humans can explain their predictions using succinct and intuitive descriptions. We show that a vision model whose feature representations are text can effectively classify ImageNet images.
  arXiv  Detail & Related papers  (2023-06-29T00:24:42Z)
• Data-driven emergence of convolutional structure in neural networks [83.4920717252233]
  We show how fully-connected neural networks solving a discrimination task can learn a convolutional structure directly from their inputs. By carefully designing data models, we show that the emergence of this pattern is triggered by the non-Gaussian, higher-order local structure of the inputs.
  arXiv  Detail & Related papers  (2022-02-01T17:11:13Z)
• Inferring halo masses with Graph Neural Networks [0.5804487044220691]
  We build a model that infers the mass of a halo given the positions, velocities, stellar masses, and radii of the galaxies it hosts. We use Graph Neural Networks (GNNs) that are designed to work with irregular and sparse data. Our model is able to constrain the masses of the halos with a $sim$0.2 dex accuracy.
  arXiv  Detail & Related papers  (2021-11-16T18:37:53Z)
• Physically Explainable CNN for SAR Image Classification [59.63879146724284]
  In this paper, we propose a novel physics guided and injected neural network for SAR image classification. The proposed framework comprises three parts: (1) generating physics guided signals using existing explainable models, (2) learning physics-aware features with physics guided network, and (3) injecting the physics-aware features adaptively to the conventional classification deep learning model for prediction. The experimental results show that our proposed method substantially improve the classification performance compared with the counterpart data-driven CNN.
  arXiv  Detail & Related papers  (2021-10-27T03:30:18Z)
• Leveraging Sparse Linear Layers for Debuggable Deep Networks [86.94586860037049]
  We show how fitting sparse linear models over learned deep feature representations can lead to more debuggable neural networks. The resulting sparse explanations can help to identify spurious correlations, explain misclassifications, and diagnose model biases in vision and language tasks.
  arXiv  Detail & Related papers  (2021-05-11T08:15:25Z)

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.