REVIEW 5 cited by
Learning to Control PDEs with Differentiable Physics
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Predicting outcomes and planning interactions with the physical world are long-standing goals for machine learning. A variety of such tasks involves continuous physical systems, which can be described by partial differential equations (PDEs) with many degrees of freedom. Existing methods that aim to control the dynamics of such systems are typically limited to relatively short time frames or a small number of interaction parameters. We present a novel hierarchical predictor-corrector scheme which enables neural networks to learn to understand and control complex nonlinear physical systems over long time frames. We propose to split the problem into two distinct tasks: planning and control. To this end, we introduce a predictor network that plans optimal trajectories and a control network that infers the corresponding control parameters. Both stages are trained end-to-end using a differentiable PDE solver. We demonstrate that our method successfully develops an understanding of complex physical systems and learns to control them for tasks involving PDEs such as the incompressible Navier-Stokes equations.
Forward citations
Cited by 5 Pith papers
-
Toward Goal-Agnostic Joint-Embedding Predictive Control of Partial Differential Equations
A goal-agnostic latent-dynamics controller for 2D Navier-Stokes improves tracking by planning against a learned kinetic-energy probe rather than raw latent-space distance.
-
Learning Physics-Guided Residual Dynamics for Deformable Object Simulation
Physics-guided residual dynamics, a spring-mass simulator plus a network that predicts velocity corrections, yields the most accurate deformable-object simulation in the paper's real-world tests.
-
Beyond Silicon: Materials, Mechanisms, and Methods for Physical Neural Computing
Physical neural substrates realize inference and adaptation via native physics and occupy complementary regimes; no single platform dominates the proposed static/dynamic benchmarks.
-
The HydroGym Reinforcement Learning Platform for Fluid Dynamics
HydroGym provides 42+ (abstract claims 61+) standardized RL environments for flow control, and reports policies that transfer across Reynolds numbers and geometries, including a 38% drag-reduction transfer claim not s...
-
Missing Physics Discovery through Fully Differentiable Finite Element-Based Machine Learning
FEML couples a differentiable finite element solver with neural networks to learn missing constitutive and thermal laws from indirect observations, with demonstrations on synthetic problems.
Discussion (0). Continue with ORCID to comment.