Shared Data and Algorithms for Deep Learning in Fundamental Physics

• URL: http://arxiv.org/abs/2107.00656v1
• Date: Thu, 1 Jul 2021 18:00:00 GMT
• Title: Shared Data and Algorithms for Deep Learning in Fundamental Physics
• Authors: Lisa Benato, Erik Buhmann, Martin Erdmann, Peter Fackeldey, Jonas Glombitza, Nikolai Hartmann, Gregor Kasieczka, William Korcari, Thomas Kuhr, Jan Steinheimer, Horst St\"ocker, Tilman Plehn and Kai Zhou
• Abstract summary: We introduce a collection of datasets from fundamental physics research -- including particle physics, astroparticle physics, and hadron- and nuclear physics. These datasets, containing had top quarks, cosmic-ray induced air showers, phase transitions in hadronic matter, and generator-level histories, are made public. We present a simple yet flexible graph-based neural network architecture that can easily be applied to a wide range of supervised learning tasks.
• Score: 4.914920952758052
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We introduce a collection of datasets from fundamental physics research -- including particle physics, astroparticle physics, and hadron- and nuclear physics -- for supervised machine learning studies. These datasets, containing hadronic top quarks, cosmic-ray induced air showers, phase transitions in hadronic matter, and generator-level histories, are made public to simplify future work on cross-disciplinary machine learning and transfer learning in fundamental physics. Based on these data, we present a simple yet flexible graph-based neural network architecture that can easily be applied to a wide range of supervised learning tasks in these domains. We show that our approach reaches performance close to state-of-the-art dedicated methods on all datasets. To simplify adaptation for various problems, we provide easy-to-follow instructions on how graph-based representations of data structures, relevant for fundamental physics, can be constructed and provide code implementations for several of them. Implementations are also provided for our proposed method and all reference algorithms.

Related papers

• Machine learning for modelling unstructured grid data in computational physics: a review [37.19820094095164]
  Unstructured grid data are essential for modelling complex geometries and dynamics in computational physics. This review is intended as a guidebook for computational scientists seeking to apply machine learning approaches to unstructured grid data. It places special focus on how ML methods can overcome the inherent limitations of traditional numerical techniques. Emerging directions such as generative models with unstructured data, reinforcement learning for mesh generation, and hybrid physics-data-driven paradigms are discussed.
  arXiv  Detail & Related papers  (2025-02-13T14:11:33Z)
• A Hybrid Virtual Element Method and Deep Learning Approach for Solving One-Dimensional Euler-Bernoulli Beams [0.0]
  A hybrid framework integrating the Virtual Element Method (VEM) with deep learning is presented. The primary aim is to explore a data-driven surrogate model capable of predicting fields across varying material displacement. A neural network architecture is introduced to separately process nodal and material-specific data, effectively capturing complex interactions.
  arXiv  Detail & Related papers  (2025-01-12T20:34:26Z)
• Conservation-informed Graph Learning for Spatiotemporal Dynamics Prediction [84.26340606752763]
  In this paper, we introduce the conservation-informed GNN (CiGNN), an end-to-end explainable learning framework. The network is designed to conform to the general symmetry conservation law via symmetry where conservative and non-conservative information passes over a multiscale space by a latent temporal marching strategy. Results demonstrate that CiGNN exhibits remarkable baseline accuracy and generalizability, and is readily applicable to learning for prediction of varioustemporal dynamics.
  arXiv  Detail & Related papers  (2024-12-30T13:55:59Z)
• The Well: a Large-Scale Collection of Diverse Physics Simulations for Machine Learning [4.861642399720316]
  Well is a large-scale collection of numerical simulations of a wide variety of physical systems. These datasets can be used individually or as part of a broader benchmark suite. We provide a unified PyTorch interface for training and evaluating models.
  arXiv  Detail & Related papers  (2024-11-30T19:42:14Z)
• The Sound of Water: Inferring Physical Properties from Pouring Liquids [85.30865788636386]
  We study the connection between audio-visual observations and the underlying physics of pouring liquids. Our objective is to automatically infer physical properties such as the liquid level, the shape and size of the container, the pouring rate and the time to fill.
  arXiv  Detail & Related papers  (2024-11-18T01:19:37Z)
• Flex: End-to-End Text-Instructed Visual Navigation with Foundation Models [59.892436892964376]
  We investigate the minimal data requirements and architectural adaptations necessary to achieve robust closed-loop performance with vision-based control policies. Our findings are synthesized in Flex (Fly-lexically), a framework that uses pre-trained Vision Language Models (VLMs) as frozen patch-wise feature extractors. We demonstrate the effectiveness of this approach on quadrotor fly-to-target tasks, where agents trained via behavior cloning successfully generalize to real-world scenes.
  arXiv  Detail & Related papers  (2024-10-16T19:59:31Z)
• Neural Embedding: Learning the Embedding of Manifold of Physics Data [5.516715115797386]
  We show that it can be a powerful step in the data analysis pipeline for many applications. We provide for the first time a viable solution to quantifying the true search capability of model search algorithms in collider physics.
  arXiv  Detail & Related papers  (2022-08-10T18:00:00Z)
• An extended physics informed neural network for preliminary analysis of parametric optimal control problems [0.0]
  We propose an extension of physics informed supervised learning strategies to parametric partial differential equations. Our main goal is to provide a physics informed learning paradigm to simulate parametrized phenomena in a small amount of time.
  arXiv  Detail & Related papers  (2021-10-26T09:39:05Z)
• An Extensible Benchmark Suite for Learning to Simulate Physical Systems [60.249111272844374]
  We introduce a set of benchmark problems to take a step towards unified benchmarks and evaluation protocols. We propose four representative physical systems, as well as a collection of both widely used classical time-based and representative data-driven methods.
  arXiv  Detail & Related papers  (2021-08-09T17:39:09Z)
• 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)
• Graph signal processing for machine learning: A review and new perspectives [57.285378618394624]
  We review a few important contributions made by GSP concepts and tools, such as graph filters and transforms, to the development of novel machine learning algorithms. We discuss exploiting data structure and relational priors, improving data and computational efficiency, and enhancing model interpretability. We provide new perspectives on future development of GSP techniques that may serve as a bridge between applied mathematics and signal processing on one side, and machine learning and network science on the other.
  arXiv  Detail & Related papers  (2020-07-31T13:21:33Z)

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.