Advances in scientific hardware, statistical algorithms, as well as easy access to databases and computing power made large volumes of high veracity materials data possible to acquire, store, and process. However, these typically large multimodal datasets conceal pertinent material property information by their size, complexity and noise. This symposium will cover a breadth of topics in material synthesis, characterization, and theory; where modern data analytics approaches play a crucial role in interpretation and discovery. An effort will be made to showcase implementation as well as highlight strengths and weaknesses of various data processing and machine learning methods in real world materials problems. The symposium will also discuss new developments in areas of data reconstruction, denoising, image processing, object tracking, correlative analysis, and data management for end use.
• Multimodal data acquisition
• Signal processing via adaptive, compressive, and dynamic sensing
• Machine learning to extract materials data
• Data fusion, image processing and object tracking
• Scalable data management and computational strategies