Goodhart's Law in Reinforcement Learning

How do we design robots and autonomous vehicles that understand the unwritten rules of human behavior? Kyle speaks with Cornell Tech professor Wendy Ju about implicit interaction, "Wizard of Oz" prototyping, and what studying pedestrians, self-driving cars, and even robotic furni ...  Show more

The data we collect about ourselves can tell us a lot—but only if the technology collecting it actually fits into our lives. Daniel Epstein explores personal informatics, from fitness trackers and food journals to baby tracking and AI, and explains why abandoning a tracking tool ...  Show more

Today, we're joined by Maohao Shen, PhD student at MIT to discuss his paper, “Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search.” We dig into how Satori leverages reinforcement learning to improve language model reasoning ...  Show more

<span style="color: #333333;">Kevin is a Computer Scientist, part-time lecturer at Paul G. Allen School of Computer Science & Engineering, and research partner at OSAM. Our discussion with Kevin revolves around:
</span>

Dr. Ehsan Haghighat is a Postdoctoral Fellow at UBC studying stochastic modeling and uncertainty quantification of engineering systems. Previously, he was a Postdoctoral Associate at MIT where he studied the assessment of induced seismicity due to CO2 sequestration and oil and ga ...  Show more

Today we’re joined by José Miguel Hernández-Lobato, a university lecturer in machine learning at the University of Cambridge. In our conversation with Miguel, we explore his work at the intersection of Bayesian learning and deep learning. We discuss how he’s been applying this to ...  Show more