Hi King Yin, The architecture looks very interesting. I am just missing the definition of the reward function (or kernel if you make it stochastic).
On the other hand, I don't understand your previous comment on the Lagrangian and Hamiltonian. I haven't seen the previous version of the paper. But you can apply an optimal control approach without having to consider the velocity at all. On Mon, Apr 29, 2019 at 3:45 PM YKY (Yan King Yin, 甄景贤) < [email protected]> wrote: > OK, revised the paper: > > > 1. Turning the environment "inwards" is a redundant idea, just the > usual reinforcement learning is OK. > 2. The Lagrangian / Hamiltonian / control theoretic sections are > completely removed... I found a serious gap in the formulation. The > Lagrangian must have partial derivatives w.r.t. the state x and the > velocity x dot, otherwise the whole apparatus of control theory cannot be > applied. It looks rather difficult to salvage this, so I'll put it back on > the shelf. > 3. The original draft was indeed rather disorganized and lack > coherence. I added some more explanation showing that it works as a whole > architecture. > 4. We planned to do some coding... but it was too rushed... if we have > a couple more days.... > > > *Artificial General Intelligence List <https://agi.topicbox.com/latest>* > / AGI / see discussions <https://agi.topicbox.com/groups/agi> + > participants <https://agi.topicbox.com/groups/agi/members> + delivery > options <https://agi.topicbox.com/groups/agi/subscription> Permalink > <https://agi.topicbox.com/groups/agi/T3cad55ae5144b323-Mc68a0583c9ce4f8e9a7e73c2> > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T3cad55ae5144b323-M5e5c926dae291a0f318b61b8 Delivery options: https://agi.topicbox.com/groups/agi/subscription
