5 ms·
It seems to me that they are basically describing a variational formulation of the "optimization perspective" of reinforcement learning, which is cool, but I am
by gajomi 6y ago
It seems to me that they are basically describing a variational formulation of the "optimization perspective" of reinforcement learning, which is cool, but I am confused... where is the supervised learning? Like what is the input and what is the output?
- bnegreve 6y agoThe way I understand it, the two subproblems are supervised in the sense that they are trained using data sampled from a fixed distribution, instead of data sampled from a distribution that changes as you update your model, as it is usually the case in RL. This makes the training more stable.
- jonnycomputer 6y agoThanks for clarifying that point.
- Cmmn_Dscndnt 6y agoIt seems more as if the authors are abusing terms from Machine Learning like "Supervised Learning".
- jonnycomputer 6y agoabusing how?