Call for Papers "Special Session on Machine Learning Surrogate Models in Science and Engineering" at IEEE ICMLA 2021

[farhad.p...@gmail.com]

Special Session on Machine Learning Surrogate Models in Science and Engineering 

IEEE International Conference on Machine Learning and Applications (ICMLA)

Website: https://www.icmla-conference.org/icmla21/index.html   

Submission Deadline: August 6, 2021 

Introduction 

Scientific observations, experiments, and simulations provide valuable data to train machine learning algorithms for constructing surrogate models, also known as metamodels, that assist in characterizing complex systems. The adoption of machine learning surrogate models holds the potential to significantly accelerate design space exploration and optimization when a closed analytical form that relates design parameters to performance criteria is unattainable. Examples of such systems abound in a wide range of scientific problems involving multiple scales, such as climate science, power grid modeling, biological simulations, structural engineering, mechanics and materials, and integrated computational materials engineering. Hence, there is a growing interest to focus on the synergistic integration of traditional methods of scientific discovery and recent machine learning techniques, e.g., neural networks. Three fundamental research directions include:

1. Assuring training data quality and adequacy under stringent resource constraints.
2. Developing efficient, reliable, and physics-informed data-driven models to fully capture the behavior of complex systems.
3. Leveraging surrogate models for performing domain-specific tasks, such as optimization, sensitivity analysis, and uncertainty quantification. 

Scope 

This special session invites submissions of original works that address the above and other unique challenges of developing machine learning surrogate models in a broad range of science and engineering applications. Topics covered by this special session include but are not limited to: 

• General methods for data acquisition, exploration, and analysis
• Active learning and optimal experimental design
• Feature extraction, selection, and dimensionality reduction
• Predictive modeling in the limited labeled data and/or weakly supervised case 
•  Machine learning methods for multimodal and heterogeneous data sources
• Scalability of machine learning algorithms
• Physics-informed machine learning
• Developing surrogate models for complex systems and model evaluation
• Exploration and optimization of design spaces with machine learning surrogate models 
• Multi-fidelity uncertainty quantification 

Special Session Organizers 

• Farhad Pourkamali-Anaraki, Assistant Professor of Computer Science, University of Massachusetts Lowell, Email: farhad_p...@uml.edu  
• Scott Stapleton, Assistant Professor of Mechanical Engineering, University of Massachusetts Lowell, Email: scott_s...@uml.edu 
• M. Amin Hariri-Ardebili,  Engineering Laboratory, National Institute of Standards and Technology (NIST), Email: amin....@nist.gov