"Hi gary,
The DGX-1 is the most powerful GPU oriented computer that exists, but you can do a lot with a much cheaper computer. In fact most people that work with deep learning use much smaller computers. With a gtx 970 you can start to do a lot of…"
"Hi john,
This is a very big project for one person, because it requires a lot more work than training a net.
The only way I can see this being developed is if a company develops the necessary framework for a competition. So that teams at…"
"That is a good point, one of advantages of hand crafted algorithms is that we can transfer our knowledge into them.
That is especially useful for events that might not be seen during training. What you can do is to manually supervise and punish…"
"Hi Hugues,
It is true that usually neural networks for object recognition are designed to output a probability distribution over all the classes -- although it can also be done other ways.
But that is not a weakness, in fact it is very useful as a…"
"Hi John,
I don't know what is your experience in deep learning. But i have been working on it full time for close to a year, i have trained dozens of nets and read hundreds of papers. And my experience tells me that this problem is perfectly…"
"Hi hugues,
The reason neural networks work so well is because they are able to generalize. For example once you have seen enough pictures of cats and you understand the "concept" o a cat, you can tell if something is a cat even when it is in a…"
"Hi Gary,
The most expensive par about deep learning is the training, and this part would be computed on servers.
The execution -- usually called inference -- is a much simpler task. The most expensive part of inference on object recognition tasks…"
A few weeks ago Chris Anderson shared a post about the efforts of bringing obstacle avoidance to PX4. This efforts are based on Simultaneous Location And Mapping (building a map of the world in real time) and Path Planning -- For the sake of…
"The asymmetric blades are more efficient in general, there is not discussion on that. The point is if the spinblades are more efficient.
I haven't been able to find any other provider of asymmetric blades, do you know if there are other ones?"