Computer Science Seminar: Alex LaGrassa (Carnegie Mellon University)

Speaker: Alex LaGrassa (Carnegie Mellon University)

Title: Robots Learning the Limits of Their Knowledge for Better Decision-Making

The seminar will be available for in-person and Zoom participation. To participate online, please email inquiry-cs@barnard.edu to receive the Zoom link.

From agriculture to manufacturing and beyond, artificial intelligence is transforming the capabilities of robots and decision-making systems, enabling them to perform tasks in a growing range of real-world scenarios. However, state-of-the-art learning methods often depend on extensive data and computational resources, making them inaccessible to many. My research focuses on developing decision-making algorithms that combine the adaptability of AI with the efficiency and reliability of classical robotics methods.

In this talk, I present an approach that enables robots to learn from experience and solve tasks despite incomplete or incorrect knowledge, using both learned and physics-based intelligence to efficiently use their resources. I demonstrate my algorithms on real-world tasks such as plant watering. This work contributes to a scalable framework for decision-making that leverages diverse types of knowledge to enable AI to expand capabilities across a wide range of resource constraints.

Alex LaGrassa (they/them) is a PhD candidate in the Robotics Institute at Carnegie Mellon University researching methods that combine machine learning with classical robotics by quantifying then expanding robot capabilities. They develop algorithms that equip robots with the intelligence to manipulate challenging but common deformable objects such as plants, cables, and liquid with limited access to data and computational resources. Alex is also passionate about inclusive STEM education so all communities contribute to shaping technological development. Towards this goal, Alex leads Paths to AI Research, a mentoring program that aims to diversify AI by introducing undergraduates from a wide range of backgrounds to research. They also enjoy designing and teaching inclusive educational material in robotics and ML.