The 2021 Nobel prize in Physics was awarded to three scientists for their groundbreaking contributions in revealing hidden patterns in complex phenomena that enhance our understanding of complex physical systems. Indeed, many real-world systems are highly intricate, nonlinear, and unpredictable. Examples include human brain, consisting of tens of billions of inter-connected neurons; the global climate system, having many nested and interlinked subsystems; a stock market composed of thousands of inter-trading companies, and a big city made up of millions of interacting people. On the one hand, the above-mentioned exemplary systems produce a large amount of data on the daily basis, which contain useful information that are yet to be explored. On the other hand, systems of a huge number of interacting units have already been studied by physicists and mathematicians using statistical mechanics and complex network theory. One excellent physical example is glass, a ubiquitous material in everyday life that is thought to be a non-equilibrium liquid. In this project, we will develop physics-guided data-driven models, such as time series analysis, machine learning, and thermodynamic framework to comprehend and further to predict the behaviors of real-world complex systems. The new models will be applied to study systems such as human brain, climate, financial systems, social networks, and physical glasses.