In design, fabrication, and control problems, we are often faced with the task of synthesis, in which we must generate an object or configuration that satisfies a set of constraints while maximizing one or more objective functions. The synthesis problem is typically characterized by a physical process in which many different realizations may achieve the goal. This many-to-one map presents challenges to the supervised learning of feed-forward synthesis, as the set of viable designs may have a complex structure. In addition, the non-differentiable nature of many physical simulations prevents efficient direct optimization. We address both of these problems with a two-stage neural network architecture that we may consider to be an autoencoder. We first learn the decoder: a differentiable surrogate that approximates the many-to-one physical realization process. We then learn the encoder, which maps from goal to design, while using the fixed decoder to evaluate the quality of the realization. We evaluate the approach on two case studies: extruder path planning in additive manufacturing and constrained soft robot inverse kinematics. We compare our approach to direct optimization of the design using the learned surrogate, and to supervised learning of the synthesis problem. We find that our approach produces higher quality solutions than supervised learning, while being competitive in quality with direct optimization, at a greatly reduced computational cost.
Uncertainty-Aware Surrogate-based Amortized Bayesian Inference for Computationally Expensive Models
Department of Stochastic Simulation and Safety Research for Hydrosystems（随机模拟与安全研究部门）；Cluster of Excellence SimTech（卓越中心SimTech）；University of Stuttgart（斯图加特大学）；Department of Statistics（统计学系）；TU Dortmund University（多特蒙德技术大学）
Amortized Neural Optimization for Pre-Layout Signal Integrity Design Space Exploration using Differentiable Surrogates
Information Processing Lab, Faculty for Electrical Engineering and Information Technology, TU Dortmund（信息处理实验室，电气工程与信息科技学院，图腾大学）；Pyramide2525, Paderborn, Germany（Pyramide2525，帕德博恩，德国）；EMC Technology Center Paderborn, Zuken GmbH, Paderborn, Germany（EMC技术中心帕德博恩，Zuken GmbH，帕德博恩，德国）
Aalto University（阿尔托大学）；Inserm（法国国家健康与医学研究院）；Bordeaux Population Health Vaccine Research Institute（波尔多人口健康疫苗研究所）；Université de Bordeaux（波尔多大学）；Inria Bordeaux Sud-ouest（法国国家信息与自动化研究所波尔多西南分部）
New directions for surrogate models and differentiable programming for High Energy Physics detector simulation
Paul Scherrer Institute（保罗·谢勒研究所）；Argonne National Laboratory（阿贡国家实验室）；SLAC National Accelerator Laboratory（SLAC国家加速器实验室）；Universität Hamburg（汉堡大学）；Rutgers University（罗格斯大学）；Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室）；Fermi National Accelerator Laboratory（费米国家加速器实验室）；Massachusetts Institute of Technology（麻省理工学院）
Fast, accurate, and differentiable: a neural-network surrogate for NRSur7dq4 precessing binary black hole waveforms
University of Rhode Island（罗德岛大学）；Institute for AI & Computational Research（人工智能与计算研究所）；California Institute of Technology（加州理工学院）；LIGO Laboratory（LIGO实验室）
Low-rank surrogate modeling and stochastic zero-order optimization for training of neural networks with black-box layers
Artificial Intelligence Research Institute (AIRI)（人工智能研究 institute）；Skolkovo Institute of Science and Technology（斯克尔科沃科学与技术研究所）；Sber Quantum Technology Center（Sber 量子技术中心）
DeepLight: A Sobolev-trained Image-to-Image Surrogate Model for Light Transport in Tissue
Chair of Biological Imaging, Central Institute for Translational Cancer Research (TranslaTUM), School of Medicine and Health & School of Computation, Information and Technology, Technical University of Munich（生物成像教授、转化癌症研究中央研究所（TranslaTUM）、医学与健康学院及计算、信息与技术学院、慕尼黑技术大学）；Institute of Biological and Medical Imaging, Bioengineering Center, Helmholtz Zentrum München（生物与医学成像研究所、生物工程中心、海德堡中心慕尼黑）；Institute of Computational Biology, Helmholtz Zentrum Munich（计算生物学研究所、海德堡中心慕尼黑）；Institute of AI for Health, Helmholtz Zentrum Munich（健康人工智能研究所、海德堡中心慕尼黑）；Ludwig-Maximilians-Universität München（路德维希-马克西米利安-慕尼黑大学）
Leveraging Recursive Gumbel-Max Trick for Approximate Inference in Combinatorial Spaces
Skolkovo Institute of Science and Technology（斯科尔科沃科学技术学院）；Airtificial Intelligence Research Institute（人工智能研究所）；Moscow Institute of Physics and Technology（莫斯科物理与技术学院）；HSE University（高等经济学院）；Skoltech（斯科尔科沃科学技术学院）；AIRI（人工智能研究所）；Sber AI Lab；MIPT（莫斯科物理与技术学院）
Physics Enhanced Deep Surrogates for the Phonon Boltzmann Transport Equation
School of Computational Science and Engineering, Georgia Institute of Technology（计算科学与工程学院，佐治亚理工学院）；Institute for Soldier Nanotechnologies, Massachusetts Institute of Technology（士兵纳米技术研究所，麻省理工学院）
NeuroPareto: Calibrated Acquisition for Costly Many-Goal Search in Vast Parameter Spaces
University of Macau（澳门大学）；Fudan University（复旦大学）；Renmin University of China（中国人民大学）；Wuhan University（武汉大学）；Shanghai AI Laboratory（上海人工智能实验室）；University of Southampton（南安普顿大学）；Hong Kong Baptist University（香港 Baptist大学）
Learned multiphysics inversion with differentiable programming and machine learning
Georgia Institute of Technology（佐治亚理工学院）；Rice University（莱斯大学）；University Medical Center Utrecht（乌得勒支大学医学中心）；Microsoft（微软）；SINTEF Digital（挪威科技工业研究院数字研究所）；Imperial College London（帝国理工学院）
