Physics-Driven ML-Based Modelling for Correcting Inverse Estimation
Not on the map. Why?
Abstract (arXiv)
When deploying machine learning estimators in science and engineering (SAE) domains, it is critical to avoid failed estimations that can have disastrous consequences, e.g., in aero engine design. This work focuses on detecting and correcting failed state estimations before adopting them in SAE inverse problems, by utilizing simulations and performance metrics guided by physical laws. We suggest to flag a machine learning estimation when its physical model error exceeds a feasible threshold, and propose a novel approach, GEESE, to correct it through optimization, aiming at delivering both low error and high efficiency. The key designs of GEESE include (1) a hybrid surrogate error model to provide fast error estimations to reduce simulation cost and to enable gradient based backpropagation of error feedback, and (2) two generative models to approximate the probability distributions of the candidate states for simulating the exploitation and exploration behaviours. All three models are constructed as neural networks. GEESE is tested on three real-world SAE inverse problems and compared to a number of state-of-the-art optimization/search approaches. Results show that it fails the least number of times in terms of finding a feasible state correction, and requires physical evaluations less frequently in general.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 25 of its 78 references that are in the map. First posted on arXiv 25 Sep 2023.
Builds on
25 of 78 references.
References · 25
- 2013Playing Atari with Deep Reinforcement Learningcited by 3,211
- 2018Soft Actor-Critic Algorithms and Applicationscited by 1,179
- 2020A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challengescited by 813
- 2017OptNet: Differentiable Optimization as a Layer in Neural Networkscited by 588
- 2017A Tutorial on Thompson Samplingcited by 420
- 2019Self-Supervised Exploration via Disagreementcited by 263
- 2018Optimization of Molecules via Deep Reinforcement Learningcited by 199
- 2020pymoo: Multi-objective Optimization in Pythoncited by 196
- 2019Conditioning by adaptive sampling for robust designcited by 110
- 2018Model-Based Active Explorationcited by 105
- 2021Conservative Objective Models for Effective Offline Model-Based Optimizationcited by 87
- 2019Model Inversion Networks for Model-Based Optimizationcited by 83
13 more
- 2015On Learning to Think: Algorithmic Information Theory for Novel Combinations of Reinforcement Learning Controllers and Recurrent Neural World Modelscited by 52
- 2021RoMA: Robust Model Adaptation for Offline Model-based Optimizationcited by 39
- 2019Momentum-Net: Fast and convergent iterative neural network for inverse problemscited by 38
- 2021Offline Model-Based Optimization via Normalized Maximum Likelihood Estimationcited by 38
- 2019Direct shape optimization through deep reinforcement learningcited by 28
- 2019Deep Neural Network Ensembles against Deception: Ensemble Diversity, Accuracy and Robustnesscited by 18
- 2022TiCo: Transformation Invariance and Covariance Contrast for Self-Supervised Visual Representation Learningcited by 12
- 2021Runtime Safety Assurance for Learning-enabled Control of Autonomous Driving Vehiclescited by 8
- 2021A General Framework Combining Generative Adversarial Networks and Mixture Density Networks for Inverse Modeling in Microstructural Materials Designcited by 4
- 2022Neural Optimization Machine: A Neural Network Approach for Optimizationcited by 4
- 2018A Deep Reinforcement Learning Approach for Dynamically Stable Inverse Kinematics of Humanoid Robotscited by 3
- 2022Global Optimization Networkscited by 2
- 2022Self-Validated Physics-Embedding Network: A General Framework for Inverse Modellingcited by 2
Built on by
Citing papers per year: 2023: 1.
1 paper.