An energy-based perspective on learning observation models


Figure 1 We show that learning observation models can be viewed as shaping energy functions that graph optimizers, even non-differentiable ones, optimize. Inference solves for most likely states given model and input measurements Learning uses training data to update observation model parameters . Robots perceive the rich world around them through the lens of their sensors. Each sensor observation is a tiny window into the world that provides only a partial, simplified view of reality. To complete their tasks, robots combine multiple readings from sensors into an internal task-specific representation of the world that we call state.

Duplicate Docs Excel Report

None found

Similar Docs  Excel Report  more

None found