Due to rapid population growth and climate change, there are severe spatiotemporal variations of water resources in the globe, and our society is facing serious challenges in securing sufficient water. To tackle the water shortage, we need to find other water sources. An effective and possible way is to utilize atmospheric water resources, which are the precipitable water in the atmosphere. With relatively stable temporal and spatial distribution, this part of water resources can be exploited and utilized through artificial precipitation enhancement operations which is also known as cloud seeding. In this talk, Professor Chen will introduce the situation of atmospheric water resources and the method that implements low frequency acoustic waves to stimulate and enhance precipitation. Through indoor experimental analysis and a large number of field tests, the effect has been tested. The development and utilization of atmospheric water resources would provide an innovative measure to obtain more freshwater for a certain region. He will also discuss the atmospheric water resources in the Greater Bay Area.
Substantial material resource recovery opportunities exist in the urban environment to support more sustainable urban development. However, the ability to produce safe and quality recoverable requires in-depth environmental materials studies and state-of-the-art fabrication and characterization technologies. For example, the quantitative X-ray diffraction (QXRD) technique has accurately monitored the transfer and behavior of targeted hazardous metals when being beneficially used for ceramic products in the construction industry. The work of recovering metallic lead from waste cathode ray tube (CRT) glass serves as an excellent example to reflect how environmental materials techniques assisted the development of transforming urban electronic waste into new metal resources. Lastly, the demonstration of recovering phosphorus from wastewater streams as quality slow-releasing fertilizer for agriculture applications leads to new solutions to tackle critical resource challenges with the fast-developing urban mining concept around the world.
Machine learning and deep neural networks have revolutionized various fields, most obvious examples are computer vision and natural language processing. Apart from the surging sizes of sophisticated models, an emerging trend is to go down the opposite route of deploying lightweight models on the edge (terminal or user end) for relatively simple AI tasks. This is named edge AI which is often constrained to run under restrictive compute and storage resources. In this talk, we will explore the latest theory in neural network modeling that allows the total avoidance of AI training that used to be slow, daunting or even impossible for the edge. Specifically, we will scratch the surface of the neural tangent kernel, and try to establish (well…. qualitatively) the equivalence of data and network, such that once the data are ready, the network is instantly ready, too.