Self-supervised Deep Reinforcement Learning with Generalized Computation Graphs for Robot Navigation
Жүктеу.....
Пікірлер: 18
@chanchoi50766 жыл бұрын
I enjoyed this.
@deeplearner26346 жыл бұрын
this is bloody awesome!
@fractelet6 жыл бұрын
good job
@primodernious5 жыл бұрын
what about use many linear neural net to works as seprarte networks but are targeted the same travel path by let each network read the same input data but compete for their output by forwarding into a output network that will select the optimum path ahed of time before the actual try? would not this speed up the overall learning performace?
@sam1712886 жыл бұрын
Hi, It was a great stuff. Anyway, a question about how you trained the agent. Is there any terminal state? For example when the agent hit the wall or flipped and how about the reward? Thank you
@gregkahn7238
6 жыл бұрын
Good question. Yes, any type of collision is a terminal state. After a collision, the car performs a hard-coded backup procedure and then continues learning. The hard-coded backup is not necessarily needed, and we are going to remove it soon. When evaluating the (N-step) Q-learning prior methods, the reward was the speed of the car, or 0 if a collision occurred. We tried adding a negative reward for collisions, but this actually hurt the performance of Q-learning.
@sam171288
6 жыл бұрын
Thank you for the reply. Can I contact you directly in case I have any further question? I am doing a Deep RL for robotic too, but a lot simple than yours.
@abhishekkumar19723 жыл бұрын
@greg kahn i will try to implement this project, any sort of help will be appreciated
@zzzzjinzj6 жыл бұрын
Can it simulate in gym-gazebo?
@gregkahn7238
6 жыл бұрын
The simulator in this release uses Bullet (for physics simulation) and Panda3d for graphics rendering. However, adding an interface to a new environment should hopefully be straightforward.
@MinhTran-ew3on4 жыл бұрын
In the paper, you claim that your approach learns from scratch to navigate using monocular images solely in the "real-world". So, did you train in a simulation and then evaluate the trained model in the real world or train it directly in the real world (it may be a real collision happen to the car to get more experience) ?
@ConsumerAria516 жыл бұрын
Do you have a paper published ? Thanks!
@gregkahn7238
6 жыл бұрын
arxiv.org/abs/1709.10489
@aitor.online6 жыл бұрын
so uc berkley isnt all bad😂 jk but this is siiick
@canislupusfool6 жыл бұрын
Fake! You can clearly hear the mouse you've trained to push it along! Nice work :)
@pranavsreedhar1402
5 жыл бұрын
dont know if you are sarcastic. Im guessing a mouse wouldn't push this fast.
Пікірлер: 18
I enjoyed this.
this is bloody awesome!
good job
what about use many linear neural net to works as seprarte networks but are targeted the same travel path by let each network read the same input data but compete for their output by forwarding into a output network that will select the optimum path ahed of time before the actual try? would not this speed up the overall learning performace?
Hi, It was a great stuff. Anyway, a question about how you trained the agent. Is there any terminal state? For example when the agent hit the wall or flipped and how about the reward? Thank you
@gregkahn7238
6 жыл бұрын
Good question. Yes, any type of collision is a terminal state. After a collision, the car performs a hard-coded backup procedure and then continues learning. The hard-coded backup is not necessarily needed, and we are going to remove it soon. When evaluating the (N-step) Q-learning prior methods, the reward was the speed of the car, or 0 if a collision occurred. We tried adding a negative reward for collisions, but this actually hurt the performance of Q-learning.
@sam171288
6 жыл бұрын
Thank you for the reply. Can I contact you directly in case I have any further question? I am doing a Deep RL for robotic too, but a lot simple than yours.
@greg kahn i will try to implement this project, any sort of help will be appreciated
Can it simulate in gym-gazebo?
@gregkahn7238
6 жыл бұрын
The simulator in this release uses Bullet (for physics simulation) and Panda3d for graphics rendering. However, adding an interface to a new environment should hopefully be straightforward.
In the paper, you claim that your approach learns from scratch to navigate using monocular images solely in the "real-world". So, did you train in a simulation and then evaluate the trained model in the real world or train it directly in the real world (it may be a real collision happen to the car to get more experience) ?
Do you have a paper published ? Thanks!
@gregkahn7238
6 жыл бұрын
arxiv.org/abs/1709.10489
so uc berkley isnt all bad😂 jk but this is siiick
Fake! You can clearly hear the mouse you've trained to push it along! Nice work :)
@pranavsreedhar1402
5 жыл бұрын
dont know if you are sarcastic. Im guessing a mouse wouldn't push this fast.
what about the code bro?