Latent Field Discovery In Interacting Dynamical Systems With Neural Fields

• URL: http://arxiv.org/abs/2310.20679v2
• Date: Wed, 20 Mar 2024 16:05:03 GMT
• Title: Latent Field Discovery In Interacting Dynamical Systems With Neural Fields
• Authors: Miltiadis Kofinas, Erik J. Bekkers, Naveen Shankar Nagaraja, Efstratios Gavves,
• Abstract summary: We focus on discovering fields, and infer them from the observed dynamics alone. We combine interactions with equivariant graph networks, and combine them with neural fields in a novel graph network that integrates field forces. Our experiments show that we can accurately discover the underlying fields in charged particles settings, traffic scenes, and gravitational n-body problems.
• Score: 34.02700914882451
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Systems of interacting objects often evolve under the influence of field effects that govern their dynamics, yet previous works have abstracted away from such effects, and assume that systems evolve in a vacuum. In this work, we focus on discovering these fields, and infer them from the observed dynamics alone, without directly observing them. We theorize the presence of latent force fields, and propose neural fields to learn them. Since the observed dynamics constitute the net effect of local object interactions and global field effects, recently popularized equivariant networks are inapplicable, as they fail to capture global information. To address this, we propose to disentangle local object interactions -- which are $\mathrm{SE}(n)$ equivariant and depend on relative states -- from external global field effects -- which depend on absolute states. We model interactions with equivariant graph networks, and combine them with neural fields in a novel graph network that integrates field forces. Our experiments show that we can accurately discover the underlying fields in charged particles settings, traffic scenes, and gravitational n-body problems, and effectively use them to learn the system and forecast future trajectories.

Related papers

• Neural force functional for non-equilibrium many-body colloidal systems [0.20971479389679337]
  We combine power functional theory and machine learning to study non-equilibrium overdamped many-body systems of colloidal particles. We first sample in steady state the one-body fields relevant for the dynamics from computer simulations of Brownian particles. A neural network is then trained with this data to represent locally in space the formally exact functional mapping from the one-body density and velocity profiles to the one-body internal force field.
  arXiv  Detail & Related papers  (2024-06-05T19:57:23Z)
• Inferring Relational Potentials in Interacting Systems [56.498417950856904]
  We propose Neural Interaction Inference with Potentials (NIIP) as an alternative approach to discover such interactions. NIIP assigns low energy to the subset of trajectories which respect the relational constraints observed. It allows trajectory manipulation, such as interchanging interaction types across separately trained models, as well as trajectory forecasting.
  arXiv  Detail & Related papers  (2023-10-23T00:44:17Z)
• Machine learning the interaction network in coupled dynamical systems [0.0]
  In a collection of interacting particles, the interaction network contains information about how various components interact with one another. In this work, we employ a self-supervised neural network model to achieve two outcomes: to recover the interaction network and to predict the dynamics of individual agents.
  arXiv  Detail & Related papers  (2023-10-05T08:29:00Z)
• Learning Heterogeneous Interaction Strengths by Trajectory Prediction with Graph Neural Network [0.0]
  We propose the attentive relational inference network (RAIN) to infer continuously weighted interaction graphs without any ground-truth interaction strengths. We show that our RAIN model with the PA mechanism accurately infers continuous interaction strengths for simulated physical systems in an unsupervised manner.
  arXiv  Detail & Related papers  (2022-08-28T09:13:33Z)
• Information is Power: Intrinsic Control via Information Capture [110.3143711650806]
  We argue that a compact and general learning objective is to minimize the entropy of the agent's state visitation estimated using a latent state-space model. This objective induces an agent to both gather information about its environment, corresponding to reducing uncertainty, and to gain control over its environment, corresponding to reducing the unpredictability of future world states.
  arXiv  Detail & Related papers  (2021-12-07T18:50:42Z)
• GINA: Neural Relational Inference From Independent Snapshots [0.0]
  We propose a graph neural network (GNN) to simultaneously learn the latent interaction graph and, conditioned on the interaction graph, the prediction of a node's observable state. GINA is based on the hypothesis that the ground truth interaction graph -- among all other potential graphs -- allows to predict the state of a node, given the states of its neighbors, with the highest accuracy.
  arXiv  Detail & Related papers  (2021-05-29T15:42:33Z)
• TRiPOD: Human Trajectory and Pose Dynamics Forecasting in the Wild [77.59069361196404]
  TRiPOD is a novel method for predicting body dynamics based on graph attentional networks. To incorporate a real-world challenge, we learn an indicator representing whether an estimated body joint is visible/invisible at each frame. Our evaluation shows that TRiPOD outperforms all prior work and state-of-the-art specifically designed for each of the trajectory and pose forecasting tasks.
  arXiv  Detail & Related papers  (2021-04-08T20:01:00Z)
• Causal Discovery in Physical Systems from Videos [123.79211190669821]
  Causal discovery is at the core of human cognition. We consider the task of causal discovery from videos in an end-to-end fashion without supervision on the ground-truth graph structure.
  arXiv  Detail & Related papers  (2020-07-01T17:29:57Z)
• 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.