Open Source Robotics
From Prediction to Consequence: A New Evaluation Paradigm for Embodied AI
Date, time, and room will be added once confirmed.
Talk overview
Modern AI excels at predicting actions from offline data, but embodied systems require evaluating how predictions affect real-world outcomes. This talk introduces a new evaluation framework that combines open-loop and closed-loop methods to assess prediction-to-consequence alignment. It uses autonomous driving and AlpaBridge, an external-driver interface, to demonstrate how policies behave differently when actions influence environment dynamics.
