Artificial Intelligence (AI) and Machine Learning (ML) have become critical tools in many scientific research domains. Particularly, in data and statistics heavy fields like particle physics, ML tools are essential to meeting the computing needs of current and future experiments and to ensuring robust data reconstruction and interpretation. This relationship is reciprocal as well as incorporating symmetries, conservation laws, and statistical methodologies from physics have led to advances in state of the art ML. In the first part of this talk I will discuss work incorporating physics-based inductive biases into ML models and ways we can begin to characterize the resulting behavior.
In addition to being powerful tools for scientific research, ML and AI are now ubiquitous in nearly all facets of society, from healthcare to criminal justice, from education to transportation. These applications have the potential to address critical community needs and improve educational, health, financial, and safety outcomes; however, they also have the potential to exacerbate existing inequalities and raise concerns about privacy, surveillance, and data ownership. In the second part of this talk I will discuss some of the unique concerns that arise when using ML to model complex systems like cities, aid networks, and political regimes and ways we can utilize scientific methodology coupled with techniques from social sciences, public health, law, and other fields, to make these models more robust.