Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userAn Extensible Benchmark Suite for Learning to Simulate Physical Systems
380
0
0.0
(
0
)
Ask ChatGPT about the research
No Arabic abstract
Simulating physical systems is a core component of scientific computing, encompassing a wide range of physical domains and applications. Recently, there has been a surge in data-driven methods to complement traditional numerical simulations methods, motivated by the opportunity to reduce computational costs and/or learn new physical models leveraging access to large collections of data. However, the diversity of problem settings and applications has led to a plethora of approaches, each one evaluated on a different setup and with different evaluation metrics. 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 integrators and representative data-driven methods (kernel-based, MLP, CNN, nearest neighbors). Our framework allows evaluating objectively and systematically the stability, accuracy, and computational efficiency of data-driven methods. Additionally, it is configurable to permit adjustments for accommodating other learning tasks and for establishing a foundation for future developments in machine learning for scientific computing.
rate research
Read More
Here we present a machine learning framework and model implementation that can learn to simulate a wide variety of challenging physical domains, involving fluids, rigid solids, and deformable materials interacting with one another. Our framework---which we term Graph Network-based Simulators (GNS)---represents the state of a physical system with particles, expressed as nodes in a graph, and computes dynamics via learned message-passing. Our results show that our model can generalize from single-timestep predictions with thousands of particles during training, to different initial conditions, thousands of timesteps, and at least an order of magnitude more particles at test time. Our model was robust to hyperparameter choices across various evaluation metrics: the main determinants of long-term performance were the number of message-passing steps, and mitigating the accumulation of error by corrupting the training data with noise. Our GNS framework advances the state-of-the-art in learned physical simulation, and holds promise for solving a wide range of complex forward and inverse problems.
Simulation is a useful tool in situations where training data for machine learning models is costly to annotate or even hard to acquire. In this work, we propose a reinforcement learning-based method for automatically adjusting the parameters of any (non-differentiable) simulator, thereby controlling the distribution of synthesized data in order to maximize the accuracy of a model trained on that data. In contrast to prior art that hand-crafts these simulation parameters or adjusts only parts of the available parameters, our approach fully controls the simulator with the actual underlying goal of maximizing accuracy, rather than mimicking the real data distribution or randomly generating a large volume of data. We find that our approach (i) quickly converges to the optimal simulation parameters in controlled experiments and (ii) can indeed discover good sets of parameters for an image rendering simulator in actual computer vision applications.
ESPResSo 4.0 is an extensible simulation package for research on soft matter. This versatile molecular dynamics program was originally developed for coarse-grained simulations of charged systems Limbach et al., Comput. Phys. Commun. 174, 704 (2006). The scope of the software has since broadened considerably: ESPResSo can now be used to simulate systems with length scales spanning from the molecular to the colloidal. Examples include, self-propelled particles in active matter, membranes in biological systems, and the aggregation of soot particles in process engineering. ESPResSo also includes solvers for hydrodynamic and electrokinetic problems, both on the continuum and on the explicit particle level. Since our last description of version 3.1 Arnold et al., Meshfree Methods for Partial Differential Equations VI, Lect. Notes Comput. Sci. Eng. 89, 1 (2013), the software has undergone considerable restructuring. The biggest change is the replacement of the Tcl scripting interface with a much more powerful Python interface. In addition, many new simulation methods have been implemented. In this article, we highlight the changes and improvements made to the interface and code, as well as the new simulation techniques that enable a user of ESPResSo 4.0 to simulate physics that is at the forefront of soft matter research.
The structural design process for buildings is time-consuming and laborious. To automate this process, structural engineers combine optimization methods with simulation tools to find an optimal design with minimal building mass subject to building regulations. However, structural engineers in practice often avoid optimization and compromise on a suboptimal design for the majority of buildings, due to the large size of the design space, the iterative nature of the optimization methods, and the slow simulation tools. In this work, we formulate the building structures as graphs and create an end-to-end pipeline that can learn to propose the optimal cross-sections of columns and beams by training together with a pre-trained differentiable structural simulator. The performance of the proposed structural designs is comparable to the ones optimized by genetic algorithm (GA), with all the constraints satisfied. The optimal structural design with the reduced the building mass can not only lower the material cost, but also decrease the carbon footprint.
This paper introduces the Behaviour Suite for Reinforcement Learning, or bsuite for short. bsuite is a collection of carefully-designed experiments that investigate core capabilities of reinforcement learning (RL) agents with two objectives. First, to collect clear, informative and scalable problems that capture key issues in the design of general and efficient learning algorithms. Second, to study agent behaviour through their performance on these shared benchmarks. To complement this effort, we open source github.com/deepmind/bsuite, which automates evaluation and analysis of any agent on bsuite. This library facilitates reproducible and accessible research on the core issues in RL, and ultimately the design of superior learning algorithms. Our code is Python, and easy to use within existing projects. We include examples with OpenAI Baselines, Dopamine as well as new reference implementations. Going forward, we hope to incorporate more excellent experiments from the research community, and commit to a periodic review of bsuite from a committee of prominent researchers.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
