AAAI-MLPS 2020
2020-03-23..2020-03-25
Stanford, United States
| title | author | length |
|---|---|---|
| Solving Inverse Problems with Deep Learning | Lexing Ying | 45m |
Neural architectures encode physics: a wavelet-based PCR-net for pseudodifferential inverses and a SwitchNet/butterfly module for Fourier integral operators… | ||
| GMLS-Nets: A Machine Learning Framework for Unstructured Data | Nathaniel Trask | 36m |
Framework generalizes convolutional networks to scattered point data by parameterizing a GMLS estimator that encodes local least-squares reconstructions into… | ||
| Enforcing Constraints for Time Series Prediction in Supervised/Unsupervised/Reinforcement Learning | Panos Stinis | 42m |
A pipeline enforces dynamical constraints to learn discrete flow maps from time series using three modes of learning and continual updates, demonstrated on the… | ||
| Nonlocal Physics-Informed Neural Networks - A Unified Framework for Nonlocal Models | Marth D'Elia | 37m |
A unified nonlocal calculus and nPINNs use a neural network surrogate for the solution (tanh activations) and optimize weights plus model parameters mu and the… | ||
| Finding Multiple Solutions of ODEs with Neural Networks | Marco Di Giovanni | 33m |
Algorithm trains multiple neural nets with a new interaction loss that penalizes pairwise closeness so different networks converge to distinct roots… | ||
| Understanding and Mitigating Gradient Flow Pathologies in PINN | Paris Perdikaris | 55m |
Diagnoses stiffness in the gradient flow of PINNs by measuring Hessian largest eigenvalues and backprop gradient histograms, showing imbalance between PDE… | ||
| Physics-Informed Spatiotemporal Deep Learning for Emulating Coupled Dynamical Systems | Diane Oyen | 36m |
A data-driven emulator uses a spatiotemporal LSTM with dual memories, gradient‑highway units and physics-informed loss terms (first and second spatial/temporal… | ||
| Physics-Informed Machine Learning for Real-time Reservoir Management | Maruti Mudunuru | 33m |
A physics-informed ML workflow combines a site behavior library built from multi-fidelity simulations (PFlowTran high‑fidelity runs plus reduced‑order Patsak… | ||
| Permeability Prediction of Porous Media using CNN with Physical Properties | Hongkyu Yoon | 23m |
Convolutional network trained on ~350 synthetic bead‑packing CT images predicts permeability by combining image features with porosity and surface area via… | ||
| Hidden Physics Models: Machine Learning of Nonlinear Partial Differential Equations | Maziar Raissi | 01h 11m |
Combines Gaussian-process surrogates and physics-informed priors to learn nonlinear PDEs and parameters from tiny datasets: design optimization of a hydrofoil… | ||
| Generalized Physics-Informed Learning through Language-Wide Differentiable Programming | Chris Rackauckas | 54m |
Language-wide automatic differentiation for full Julia programs generates gradients of real scientific code so existing solvers become trainable: Zygote… | ||
| DeepXDE: A Deep Learning Library for Solving Differential Equations | Lu Lu | 48m |
DeepXDE is a Python library that implements physics‑informed neural networks using TensorFlow and automatic differentiation to embed PDEs (Laplacian… | ||
| Data-Driven Inverse Modeling with Incomplete Observations | Kailai Xu | 32m |
Presents a method that trains neural nets while enforcing PDE constraints exactly by making both explicit and implicit numerical solvers differentiable with… | ||
| On Human-robot Interactions via Multimodal Intent Modeling and Reachability-based Safety Assurance | Marco Pavone | 59m |
Presents a decision stack that learns a conditional multimodal model of human driving with a recurrent conditional VAE, builds a library of ~4,000 robot action… | ||
| 2D CNN-based Nearshore Bathymetry Inversion from Synthetic Imagery | Adam Collins | 25m |
A 2D FCN learns to map time-averaged "timex" images from the surf zone to depth using photorealistic synthetic imagery generated by Celeris, trained with a 2D… | ||
| Deep Sensing of Ocean Wave Heights with Synthetic Aperture Radar | Brandon Quach | 27m |
Curates 2015–2018 co-locations between Sentinel-1 SAR and altimeters to train a convolutional neural network on 2D image modulation spectra (real+imag after 2D… | ||
| Surfzone Topography-informed Deep Learning Techniques to Nearshore Bathymetry with Sparse Data | Yizhou Qian | 25m |
Uses a conditional GAN trained on FRF Duck NC surveys (240 of 265 surveys, ~10,000 augmented samples) to interpolate nearshore bathymetry from sparse point… | ||
| Bayesian Optimization and Machine Learning for Accelerating Experiments in the Physical Sciences | Stefano Ermon | 01h 10m |
Describes a practical pipeline that recalibrates confidence intervals for regression and classification by learning a mapping from predicted probabilities to… | ||
| A Weighted Sparse-Input NN Applied to Identify Important Features for Vortex-Induced Vibration | Leixin Ma | 28m |
Proposes a WSPINN weighted sparse-input neural network that applies group sparsity on the input first layer to pick physics features from experiments on a 38 m… | ||
| Continuous Representation Of Molecules using Graph Variational Autoencoder | Mohammadamin Tavakoli | 38m |
Proposes a graph VAE on 2D adjacency tensor that encodes molecules with stacked GCNs (three layers, first two weights shared), trains on ZINC molecules… | ||
| Nonlinear Model Reduction: Using ML to Enable Rapid Simulation of Extreme-Scale Physics Models | Kevin Carlberg | 01h 04m |
Uses convolutional autoencoders and LSTM error models to learn a low‑dimensional nonlinear manifold (decoder = parameterization) and solves reduced dynamics… | ||
| nothing matches | ||