ML4Eng @ NeurIPS 2020
For questions, issues, or other help on the day, please email:
Link for all livestreamed sessions and on-demand video presentations of the accepted papers:
Link for poster sessions and break in


Modern engineering is driven by computational workflows for modeling, simulation, and design. Machine learning methods show promise for augmenting engineering modeling, simulation and design across many application areas. Success stories from recent research include predicting the behavior of materials, using ML for physical simulation and design optimization, creating generative models of 3d objects and designs, optimizing energy systems to improve performance and reduce environmental impact, and many more.

This workshop aims to bring together the machine learning and engineering communities to foster collaboration and spur new research in the emerging field at their intersection. Broadly speaking, we are interested in work that develops or investigates machine learning methdos for application to engineering problems. Some ideas of specific topics we want to see addressed include, but are not limited to:

  • Learning surrogate and reduced-order models for physical simulators,
  • Using machine learning to accelerate design optimization, system identification and control,
  • Developing ML tools which can assist engineers to specify, explore and evaluate possible systems and designs,
  • Using ML to accelerate prototyping and manufacturing,
  • Integrating ML tools with preexisting engineering tools and workflows,
  • Development of software and libraries to enable all of the above,
  • Identification of applications and pressing challenges/opportunities in the application of ML to engineering.


Time (EST)Session
7.50am-8.00am: Opening remarks
8.00am-8.40am: Invited talk: Nils Thuerey - Lead the way! Deep learning via differentiable simulations
8.40am-9.20am: Invited talk: Angela Dai - Self-supervised generation of 3D shapes and scenes
9.20am-11.20am: Poster session 1 on
11.20am-12.00pm: Invited talk: Tatiana Lopez-Guevara - Robots, Liquids & Inference
12.00pm-12.40pm: Invited talk: Peter Battaglia - Structured models of physics, objects, and scenes
12.40pm-1.30pm: Break
1.30pm-2.30pm: Panel discussion
2.30pm-3.10pm: Invited talk: Karen E. Willcox - Operator inference: bridging model reduction and scientific machine learning
3.10pm-3.50pm: Invited talk: Grace X. Gu - Artificial intelligence for materials design and additive manufacturing
3.50pm-4.00pm: Closing remarks
4.00pm-6.00pm: Poster session 2 on

Accepted Papers

Poster Session 1: 9.20am EST - 11.20am EST

01: Continuous calibration of a digital twin; a particle filter approach
Rebecca Ward (The Alan Turing Institute); Ruchi Choudhary (The Alan Turing Institute); Alastair Gregory (The Alan Turing Institute); Mark Girolami (
02: An Industrial Application of Deep Reinforcement Learning for Chemical Production Scheduling
Christian D Hubbs (Dow Chemical); Adam Kelloway (Dow); John Wassick (Dow Chemical); Nikolaos Sahinidis (Georgia Tech); Ignacio Grossmann (Carnegie Mellon University)
03: TPINN: An improved architecture for distributed physics informed neural networks
Sreehari M (IIT MADRAS); Balaji Srinivasan (IIT MADRAS)
04: Multilevel Delayed Acceptance MCMC with an Adaptive Error Model in PyMC3
Mikkel B Lykkegaard (University of Exeter); Greg Mingas (The Alan Turing Institute); Robert Scheichl (Heidelberg University); Colin Fox (University of Otago); Tim Dodwell (University of Exeter)
05: Machine Learning-based Anomaly Detection with Magnetic Data
Peetak Mitra (UMass Amherst); Denis Akhiyarov (Total); Mauricio Araya-Polo (Total E&P RT); Daniel Byrd (Total E&P RT)
06: Electric Vehicle Range Improvement by Utilizing Deep Learning to Optimize Occupant Thermal Comfort
Alok Warey (General Motors Global Research and Development); Shailendra Kaushik (General Motors Global Research and Development); Bahram Khalighi (General Motors Global Research and Development); Michael Cruse (Siemens Digital Industries Software); Ganesh Venkatesan (Siemens Digital Industries Software)
07: On the Effectiveness of Bayesian AutoML methods for Physics Emulators
Peetak Mitra (UMass Amherst); Niccolo Dal Santo (MathWorks, Inc.); Majid Haghshenas (UMass Amherst); Shounak Mitra (MathWorks, Inc.); Conor Daly (MathWorks, Inc.); David Schmidt (University of Massachusetts)
08: Data-driven inverse design optimization of magnetically programmed soft structures
Alp C Karacakol (Carnegie Mellon University); Yunus Alapan (Max Planck Institute for Intelligent Systems); Metin Sitti (Carnegie Mellon University)
09: Parameterized Reinforcement Learning for Optical System Optimization
Heribert Wankerl (OSRAM Opto Semiconductors GmbH); Maike Stern (OSRAM Opto Semiconductors GmbH); Ali Mahdavi (OSRAM Opto Semiconductors GmbH); Christoph Eichler (OSRAM Opto Semiconductors GmbH); Elmar Lang (University of Regensburg)
10: Efficient nonlinear manifold reduced order model
Youngkyu Kim (Mechanical Engineering, University of California, Berkeley); Youngsoo Choi (Center for Applied Scientific Computing, Lawrence Livermore National Laboratory); David Widemann (Computational Engineering Division, Lawrence Livermore National Laboratory); Tarek Zohdi (Mechanical Engineering, University of California, Berkeley)
11: Predicting Nanorobot Shapes via Generative Models
Emma Benjaminson (Carnegie Mellon University); Rebecca Taylor (Carnegie Mellon University); Matthew Travers (CMU)
12: Analog Circuit Design with Dyna-Style Reinforcement Learning
Wook Lee (Delft University of Technology); Frans Oliehoek (Delft University of Technology)
13: Surrogates for Stiff Nonlinear Systems using Continuous Time Echo State Networks
Ranjan Anantharaman (MIT); Christopher V Rackauckas (Massachusetts Institute of Technology); Viral Shah (Julia Computing)
14: Learning Mesh-Based Simulation with Graph Networks
Tobias Pfaff (DeepMind); Meire Fortunato (DeepMind); Alvaro Sanchez-Gonzalez (DeepMind); Peter Battaglia (DeepMind)
15: Learning Partially Known Stochastic Dynamics with Empirical PAC Bayes
Manuel Haußmann (Heidelberg University); Sebastian Gerwinn (Bosch Center for Artificial Intelligence); Andreas Look (Bosch); Barbara Rakitsch (Bosch Center for Artificial Intelligence); Melih Kandemir (Bosch Center for Artificial Intelligence)
16: A Sequential Modelling Approach for Indoor Temperature Prediction and Heating Control in Smart Buildings
Yongchao Huang (University of Oxford); Hugh Miles (Atamate Ltd.); Pengfei Zhang (Institute for Inforcomm Research)
17: Simultaneous Process Design and Control Optimization using Reinforcement Learning
Steven Sachio (Imperial College London); Antonio del Rio Chanona (Imperial College London); Panagiotis Petsagkourakis (University College London)
18: Accelerating Inverse Design of Nanostructures Using Manifold Learning
Mohammadreza Zandehshahvar (Georgia Institute of Technology); Yashar Kiarashinejad (Georgia Institute of Technology); Muliang Zhu (Georgia Institute of Technology); Hossein Maleki (Georgia Tech); Omid Hemmatyar (Georgia Tech); Sajjad Abdollahramezani (Georgia Tech); Reza Pourabolghasem (Independent Researcher); Ali Adibi (Georgia Tech)
19: Uncertainty-aware Remaining Useful Life predictors
Luca Biggio (ETH Zürich); Manuel Arias Chao (ETH Zurich); Olga Fink (ETH Zurich)
20: Model Order Reduction using a Deep Orthogonal Decomposition
Daniel Tait (University of Warwick)
21: Scalable Multitask Latent Force Models with Applications to Predicting Lithium-ion Concentration
Daniel Tait (University of Warwick); Ferran Brosa-Planella (University of Warwick); Widanalage Dhammika Widanage (University of Warwick); Theodoros Damoulas (University of Warwick)
22: Learning to Identify Drilling Defects in Turbine Blades with Single Stage Detectors
Andrea Panizza (Baker Hughes); Szymon Tomasz Stefanek (Siena Artificial Intelligence Laboratory); Stefano Melacci (University of Siena); Giacomo Veneri (Baker Hughes); Marco Gori (University of Siena)
23: Constraint active search for experimental design
Gustavo Malkomes (SigOpt); Bolong Cheng (SigOpt); Mike Mccourt (SigOpt)
24: Exact Preimages of Neural Network Aircraft Collision Avoidance Systems
Kyle Matoba (EPFL); François Fleuret (University of Geneva)
25: Probabilistic Adjoint Sensitivity Analysis for Fast Calibration of Partial Differential Equation Models
Jon Cockayne (Alan Turing Institute); Andrew Duncan (Imperial College London)
26: Combinatorial 3D Shape Generation via Sequential Assembly
Jungtaek Kim (POSTECH); Hyunsoo Chung (POSTECH); Jinhwi Lee (POSTECH); Minsu Cho (POSTECH); Jaesik Park (POSTECH)
27: Multi-Loss Sub-Ensembles for Accurate Classification with Uncertainty Estimation
Omer Achrack (Intel); Raizy Kellerman (Intel); Ouriel Barzilay (Intel Corporation)
28: A data centric approach to generative modelling of rough surfaces: An application to 3D-printed Stainless Steel
Liam Fleming (Newcastle University)
29: Battery Model Calibration with Deep Reinforcement Learning
Ajaykumar Unagar (ETH Zurich); Yuan Tian (ETH); Manuel Arias Chao (ETH Zurich); Olga Fink (ETH Zurich)
30: Robotic gripper design with Evolutionary Strategies and Graph Element Networks
Ferran Alet (MIT); Maria Bauza Villalonga (MIT); Adarsh Keshav S Jeewajee (Massachusetts Institute of Technology); Max Thomsen (MIT); Alberto Rodriguez (MIT); Leslie Kaelbling (MIT); Tomas Lozano-Perez (MIT)

Poster Session 2: 4pm EST - 6pm EST

31: A General Framework Combining Generative Adversarial Networks and Mixture Density Networks for Inverse Modeling in Microstructural Materials Design
Zijiang Yang (Northwestern University); Dipendra Jha (Northwestern University); Arindam Paul (Northwestern University); Wei-keng Liao (Northwestern University); Alok Choudhary (Northwestern University); Ankit Agrawal (Northwestern University)
32: Information-Theoretic Multi-Objective Bayesian Optimization with Continuous Approximations
Syrine Belakaria (Washington State university); Aryan Deshwal (Washington state university); Janardhan Rao Doppa (Washington State University)
33: Bayesian polynomial chaos
Pranay Seshadri (Imperial College London); Andrew Duncan (Imperial College London); Ashley Scillitoe (The Alan Turing Institute)
34: Scalable Combinatorial Bayesian Optimization with Tractable Statistical models
Aryan Deshwal (Washington state university); Syrine Belakaria (Washington State university); Janardhan Rao Doppa (Washington State University)
35: End-to-End Differentiability and Tensor Processing Unit Computing to Accelerate Materials’ Inverse Design
HAN LIU (University of California, Los Angeles); Yuhan Liu (University of California, Los Angeles); Zhangji Zhao (University of California, Los Angeles); Samuel S Schoenholz (Google Brain); Ekin D Cubuk (Google Brain); Mathieu Bauchy (University of California, Los Angeles)
36: Placement in Integrated Circuits using Cyclic Reinforcement Learning and Simulated Annealing
Dhruv Vashisht (Carnegie Mellon University); Harshit Rampal (Carnegie Mellon University); Haiguang Liao (Carnegie Mellon University); Yang Lu (Cadence Design Systems, Inc.); Devika Shanbhag (Carnegie Mellon University); Elias Fallon (Cadence Design Systems, Inc.); Levent Burak Kara (Carnegie Mellon University)
37: Context-Aware Urban Energy Efficiency Optimization Using Hybrid Physical Models
Benjamin Choi (Stanford University); Alex Nutkiewicz (Stanford University); Rishee Jain (Stanford University)
38: Decoding the genome of cement by Gaussian Process Regression
Yu Song (University of California, Los Angeles); Yongzhe Wang (University of California, Los Angeles); Kaixin Wang (University of California, Los Angeles); Mathieu Bauchy (University of California, Los Angeles)
39: Efficient Nanopore Optimization by CNN-accelerated Deep Reinforcement Learning
Yuyang Wang (Carnegie Mellon University); Zhonglin Cao (Carnegie Mellon University); Amir Barati Farimani (Carnegie Mellon University)
40: Frequency-compensated PINNs for Fluid-dynamics Design Problems
Tongtao Zhang (Siemens Corporation); Biswadip Dey (Siemens); Pratik Kakkar (Siemens); Arindam Dasgupta (Siemens); Chakraborty Amit (Siemens)
41: Flaw Detection in Metal Additive Manufacturing Using Deep Learned Acoustic Features
Wentai Zhang (CARNEGIE MELLON UNIVERSITY); Brandon Abranovic (CARNEGIE MELLON UNIVERSITY); Jacob Hanson-Regalado (University of California, Berkeley); Can Koz ( University of Southern California); Bhavya Duvvuri (CARNEGIE MELLON UNIVERSITY); Kenji Shimada (Carnegie Mellon University); Jack Beuth (CARNEGIE MELLON UNIVERSITY); Levent Burak Kara (Carnegie Mellon University)
42: Prediction of high frequency resistance in polymer electrolyte membrane fuel cells using Long Short Term Memory based model
Tong Lin (Carnegie Mellon University); Leiming Hu (Carnegie Mellon University); Willetta Wisely (Carnegie Mellon University); Xin Gu (Shanghai Hydrogen Propulsion Technology Co., Ltd.); Jun Cai (Shanghai Hydrogen Propulsion Technology Co., Ltd.); Shawn Litster (Carnegie Mellon University); Levent Burak Kara (Carnegie Mellon University)
43: Differentiable Implicit Layers
Andreas Look (Bosch); Simona E. Doneva (University of Mannheim); Melih Kandemir (Bosch Center for Artificial Intelligence); Rainer Gemulla (Universität Mannheim); Jan Peters (TU Darmstadt + Max Planck Institute for Intelligent Systems)
44: Rethink AI-based Power Grid Control: Diving Into Algorithm Design
Xiren Zhou (GEIRINA); siqi wang (GEIRINA); Ruisheng Diao (GEIRINA); Desong Bian (GEIRI North America); Jiajun Duan (GEIRI North America); Di Shi (GEIRI North America)
45: A Learning-boosted Quasi-Newton Method for AC Optimal Power Flow
Kyri Baker (University of Colorado, Boulder)
46: On Training Effective Reinforcement Learning Agents for Real-time Power Grid Operation and Control
Ruisheng Diao (GEIRI North America); Di Shi (GEIRI North America); Bei Zhang (GEIRI North America); Siqi Wang (GEIRI North America); Haifeng Li (SGCC Jiangsu Electric Power Company); Chunlei Xu (SGCC Jiangsu Electric Power Company); Tu Lan (GEIRI North America); Desong Bian (GEIRI North America); Jiajun Duan (GEIRI North America)
47: Jacobian of Generative Models for Sensitivity Analysis of Photovoltaic Device Processes
Maryam Molamohammadi (McGill University); Sahand Rezaei-Shoshtari (McGill University ); Nathaniel Quitoriano (McGill University)
48: Signal Enhancement for Magnetic Navigation Challenge Problem
Albert Gnadt (MIT); Joseph Belarge (MIT Lincoln Laboratory); Aaron Canciani (AFIT); Lauren Conger (MIT Lincoln Laboratory); Joseph Curro (AFIT); Alan Edelman (MIT); David Jacobs (Department of the Air Force AI Accelerator); Peter Morales (Microsoft); Michael F O\'Keeffe (MIT Lincoln Laboratory); Jonathan Taylor (MIT Lincoln Laboratory); Christopher Rackauckas (MIT)
49: Modular mobile robot design selection with deep reinforcement learning
Julian Whitman (Carnegie Mellon University); Matthew Travers (CMU); Howie Choset (Carnegie Melon University)
50: Collaborative Multidisciplinary Design Optimization with Neural Networks
Jean de Becdelievre (Stanford University); Ilan Kroo (Stanford University)
51: Heat risk assessment using surrogate model for meso-scale surface temperature
Byeongseong Choi (Carnegie Mellon University); Matteo Pozzi (Carnegie Mellon University); Mario Bergés (CMU)
52: ManufacturingNet: A Machine Learning Toolbox for Engineers
Rishikesh Magar (Carnegie Mellon University); Lalit Ghule (Carnegie Mellon University); Ruchit S Doshi (Carnegie Mellon University); Aman Khalid (UCSD); Sharan Seshadri (Carnegie Mellon University); Amir Barati Farimani (Carnegie Mellon University)
53: An adversarially robust approach to security-constrained optimal power flow
Neeraj Vijay Bedmutha (Carnegie Mellon University); Priya L Donti (Carnegie Mellon University); Zico Kolter (Carnegie Mellon University)
54: Real-time Prediction of Soft Tissue Deformations Using Data-driven Nonlinear Presurgical Simulations
Haolin Liu (Carnegie Mellon University); Ye Han (Carnegie Mellon University); Daniel J Emerson (Carnegie Mellon University); Houriyeh Majditehran (Carnegie Mellon University); Qi Wang (Carnegie Mellon University); Yoed Rabin (Carnegie Mellon University); Levent Burak Kara (Carnegie Mellon University)
55: Building LEGO using Deep Generative Models of Graphs
Rylee Thompson (University of Guelph); Graham Taylor (University of Guelph); Terrance DeVries (University of Guelph); Elahe Ghalebi (Vector & UofG)
56: Multi-Stage Transmission Line Flow Control Using Centralized and Decentralized Reinforcement Learning Agents
Xiumin Shang (University of California, Merced); Jingping Yang (Jinhua Electric Power Company); Bingquan Zhu (SGCC Zhejiang Electric Power Company); Lin Ye (SGCC Zhejiang Electric Power Company); Jing Zhang (SGCC Zhejiang Electric Power Company); Jianping Xu (Jinhua Electric Power Company); Qin Lyu (Jinhua Electric Power Company); Ruisheng Diao (GEIRI North America)
57: A Nonlocal-Gradient Descent Method for Inverse Design in Nanophotonics
Sirui Bi (Johns Hopkins University); Jiaxin Zhang (Oak Ridge National Laboratory); Guannan Zhang (Oak Ridge National Laboratory)
58: Autonomous Control of a Particle Accelerator using Deep Reinforcement Learning
Xiaoying Pang (Apple); Sunil Thulasidasan (Los Alamos National Laboratory); Larry Rybarcyk (Los Alamos National Laboratory)
59: Scalable Deep-Learning-Accelerated Topology Optimization for Additively Manufactured Materials
Sirui Bi (Johns Hopkins University); Jiaxin Zhang (Oak Ridge National Laboratory); Guannan Zhang (Oak Ridge National Laboratory)