ANONYMOUS wrote There is not a clear answer, since as I say In a system with no known model where we are not able to assume anything about the structure of the system, they become basically equivalent. This information is covered in the lectures. In g...

so in an unknown environment, which one should be used? Q-learning can just learn state-action pairs, but utility learning can just learn the transition model then use it to calculate state utilities. I'm having difficulty finding pros and cons of usi...

As was discussed in the lectures The upside of Q-learning is that it learns the transition model as part of the (state, action) utilities (where regular learning requires us to either know the transition model or learn it separately). The primary down...

I can't seem to understand the difference between q-learning and utility learning clearly, I know that q learning means learning the utility of actions available to a state rather than learning the utility of each state(util learning). But when it com...

Tim has kindly re-recorded his lecture and uploaded it.

Hi there, I think this weeks lecture by Dr Tim French has no audio. I understand some of the content in the unit is new, but are there any previous recordings of the content available? Thanks, Sam