Modeling Oceanic Variables with Dynamic Graph Neural Networks
- URL: http://arxiv.org/abs/2206.12746v1
- Date: Sat, 25 Jun 2022 22:43:02 GMT
- Title: Modeling Oceanic Variables with Dynamic Graph Neural Networks
- Authors: Caio F. D. Netto, Marcel R. de Barros, Jefferson F. Coelho, Lucas P.
de Freitas, Felipe M. Moreno, Marlon S. Mathias, Marcelo Dottori, F\'abio G.
Cozman, Anna H. R. Costa, Edson S. Gomi, Eduardo A. Tannuri
- Abstract summary: We describe a data-driven method to predict environmental variables in the region of Santos-Sao Vicente-Bertioga Estuarine System in Brazil.
Our model exploits both temporal and spatial inductive biases by joining state-of-the-art sequence models and relational models.
Experiments show that better results are attained by our model, while maintaining flexibility and little domain knowledge dependency.
- Score: 0.09830751917335563
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Researchers typically resort to numerical methods to understand and predict
ocean dynamics, a key task in mastering environmental phenomena. Such methods
may not be suitable in scenarios where the topographic map is complex,
knowledge about the underlying processes is incomplete, or the application is
time critical. On the other hand, if ocean dynamics are observed, they can be
exploited by recent machine learning methods. In this paper we describe a
data-driven method to predict environmental variables such as current velocity
and sea surface height in the region of Santos-Sao Vicente-Bertioga Estuarine
System in the southeastern coast of Brazil. Our model exploits both temporal
and spatial inductive biases by joining state-of-the-art sequence models (LSTM
and Transformers) and relational models (Graph Neural Networks) in an
end-to-end framework that learns both the temporal features and the spatial
relationship shared among observation sites. We compare our results with the
Santos Operational Forecasting System (SOFS). Experiments show that better
results are attained by our model, while maintaining flexibility and little
domain knowledge dependency.
Related papers
- Coupled Ocean-Atmosphere Dynamics in a Machine Learning Earth System Model [0.6008008212472723]
We present the Ocean-linked-atmosphere (Ola) model, a high-resolution (0.25deg) Artificial Intelligence/ Machine Learning (AI/ML) coupled earth-system model.
We find that Ola exhibits learned characteristics of ocean-atmosphere coupled dynamics including tropical oceanic waves with appropriate phase speeds.
We present initial evidence of skill in forecasting the El Nino/Southern Oscillation (ENSO) which compares favorably to the SPEAR model of the Geophysical Fluid Dynamics Laboratory.
arXiv Detail & Related papers (2024-06-12T20:29:14Z) - Modeling Randomly Observed Spatiotemporal Dynamical Systems [7.381752536547389]
Currently available neural network-based modeling approaches fall short when faced with data collected randomly over time and space.
In response, we developed a new method that effectively handles such randomly sampled data.
Our model integrates techniques from amortized variational inference, neural differential equations, neural point processes, and implicit neural representations to predict both the dynamics of the system and the timings and locations of future observations.
arXiv Detail & Related papers (2024-06-01T09:03:32Z) - Spatiotemporal Implicit Neural Representation as a Generalized Traffic Data Learner [46.866240648471894]
Spatiotemporal Traffic Data (STTD) measures the complex dynamical behaviors of the multiscale transportation system.
We present a novel paradigm to address the STTD learning problem by parameterizing STTD as an implicit neural representation.
We validate its effectiveness through extensive experiments in real-world scenarios, showcasing applications from corridor to network scales.
arXiv Detail & Related papers (2024-05-06T06:23:06Z) - Spatio-Temporal Branching for Motion Prediction using Motion Increments [55.68088298632865]
Human motion prediction (HMP) has emerged as a popular research topic due to its diverse applications.
Traditional methods rely on hand-crafted features and machine learning techniques.
We propose a noveltemporal-temporal branching network using incremental information for HMP.
arXiv Detail & Related papers (2023-08-02T12:04:28Z) - Towards Spatio-temporal Sea Surface Temperature Forecasting via Static
and Dynamic Learnable Personalized Graph Convolution Network [9.189893653029076]
This paper proposes a novel static and dynamic learnable personalized graph convolution network (SD-LPGC)
Specifically, two graph learning layers are first constructed to respectively model the stable long-term and short-term evolutionary patterns hidden in the SST signals.
Then, a learnable personalized convolution layer is designed to fuse this information.
arXiv Detail & Related papers (2023-04-12T14:35:38Z) - TempSAL -- Uncovering Temporal Information for Deep Saliency Prediction [64.63645677568384]
We introduce a novel saliency prediction model that learns to output saliency maps in sequential time intervals.
Our approach locally modulates the saliency predictions by combining the learned temporal maps.
Our code will be publicly available on GitHub.
arXiv Detail & Related papers (2023-01-05T22:10:16Z) - Temporal Relevance Analysis for Video Action Models [70.39411261685963]
We first propose a new approach to quantify the temporal relationships between frames captured by CNN-based action models.
We then conduct comprehensive experiments and in-depth analysis to provide a better understanding of how temporal modeling is affected.
arXiv Detail & Related papers (2022-04-25T19:06:48Z) - How Well Do Sparse Imagenet Models Transfer? [75.98123173154605]
Transfer learning is a classic paradigm by which models pretrained on large "upstream" datasets are adapted to yield good results on "downstream" datasets.
In this work, we perform an in-depth investigation of this phenomenon in the context of convolutional neural networks (CNNs) trained on the ImageNet dataset.
We show that sparse models can match or even outperform the transfer performance of dense models, even at high sparsities.
arXiv Detail & Related papers (2021-11-26T11:58:51Z) - SPATE-GAN: Improved Generative Modeling of Dynamic Spatio-Temporal
Patterns with an Autoregressive Embedding Loss [4.504870356809408]
We propose a novel loss objective combined with -GAN based on an autogressive embedding to reinforce the learning oftemporal dynamics.
We show that our embedding loss improves performance without any changes to the architecture of -GAN, highlighting our model's increased capacity for autocorrelationre structures.
arXiv Detail & Related papers (2021-09-30T12:10:05Z) - Causal Navigation by Continuous-time Neural Networks [108.84958284162857]
We propose a theoretical and experimental framework for learning causal representations using continuous-time neural networks.
We evaluate our method in the context of visual-control learning of drones over a series of complex tasks.
arXiv Detail & Related papers (2021-06-15T17:45:32Z)
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.