Machine Learning and Friends Lunch
Machine Learning and Friends Lunch: David Burt, Consistent Validation for Predictive Methods in Spatial Settings
David Burt is a postdoc in Professor Tamara Broderick’s group at the MIT
Laboratory For Information and Decision Systems.
Machine Learning and Friends Lunch: David Held, Relational Learning for Robot Manipulation
David Held is an Associate Professor at Carnegie Mellon University in the Robotics Institute and is the director of the RPAD lab: Robots Perceiving And Doing.
Machine Learning and Friends Lunch: Karin de Langis, Artificial Cognition in LLMs
Karin de Langis, PhD candidate at University of Minnesota, studies artificial cognition in LLMs, explaining failures and comparing cognitive control vs humans.
Machine Learning and Friends Lunch: Andrew Lee, Decomposing Query-Key Feature Interactions Using Contrastive Covariances
Analyze Transformer attention via QK space, decomposing it into low-rank, interpretable features that explain why tokens attend and how attention scores arise.
Machine Learning and Friends Lunch: Kuan Fang, Open-World Robot Dexterity via Physically Grounded Reasoning
Robot dexterity needs more than scale: I’ll show how structured affordance, contact, and motion reasoning links foundation models to open-world control.