Periodic Reporting for period 1 - SpaTe (Spatio-Temporal Methods for Data-driven Computer Animation and Simulation)
Période du rapport: 2020-09-01 au 2022-02-28
This grant project aims for a fusion of space-time physics with machine learning algorithms that will allow us to fundamentally improve the way we work with computer simulations, and benefit forward as well as inverse problem solvers with physical constraints. More informally, the goal of "SpaTe" is to develop new classes of data-driven spatio-temporal methods and demonstrate their potential to drive the next generation of numerical simulations.
It consists of three parallel and synergetic lines of work:
1) Target learning generic and re-usable representations based on neural networks that target grid-based space-time functions of physical problems.
2) Employ advanced numerical techniques for discretizing the differential operators of model equations to arrive at robust, unsupervised learning algorithms for physical phenomena.
3) Develop adaptive algorithms for sparse, point-based space-time functions to analyze and disambiguate complex data sets such as point clouds without correspondences.
It consists of three parallel and synergetic lines of work:
1) Target learning generic and re-usable representations based on neural networks that target grid-based space-time functions of physical problems.
2) Employ advanced numerical techniques for discretizing the differential operators of model equations to arrive at robust, unsupervised learning algorithms for physical phenomena.
3) Develop adaptive algorithms for sparse, point-based space-time functions to analyze and disambiguate complex data sets such as point clouds without correspondences.
In the first period of this project, we have focused on milestones 1.A 2.A and 3.A of the grant proposal. More specifically, we have worked on improving reduced temporal representations (1.A) new algorithms for physics based inverse problems solvers (2.A) and improved learning for Lagrangian representations (3.A).
While several publications for all three areas are in the works, so far only two of them have been published:
one of them targets the physical reconstruction of smoke phenomena with physical learning, and has been successfully published at the renowned CVPR conference. The other paper proposed the half-inversion of gradients for deep learning, and was published at ICLR.
While we see both publications as success stories, we expect the total number of publications to increase significantly in the following periods. Several ongoing works were delayed due to particularly (and in our opinion unnecessarily) unfavourable reviews.
While several publications for all three areas are in the works, so far only two of them have been published:
one of them targets the physical reconstruction of smoke phenomena with physical learning, and has been successfully published at the renowned CVPR conference. The other paper proposed the half-inversion of gradients for deep learning, and was published at ICLR.
While we see both publications as success stories, we expect the total number of publications to increase significantly in the following periods. Several ongoing works were delayed due to particularly (and in our opinion unnecessarily) unfavourable reviews.
Both projects mentioned above have successfully improved the state of the art in their respective areas.
In the longer term, we expect that this project will allow us to better understand the physical world around us. It will help us to analyze sparse and ambiguous mea- surements such as videos and 3D scans automatically and reliably, with a vast range of practical applications from social-media apps to autonomous vehicles.
In the longer term, we expect that this project will allow us to better understand the physical world around us. It will help us to analyze sparse and ambiguous mea- surements such as videos and 3D scans automatically and reliably, with a vast range of practical applications from social-media apps to autonomous vehicles.