deepnet

62 ms·

631.
by deepnet 11y ago
My, lightsail is yar. Easy to handle, quick to the helm, fast, bright, everything a ship should be.
632.
by deepnet 11y ago
This is very challenging. Very very challenging
633.
by deepnet 11y ago
Is there a recent md5 of the current state of chain world ? I would like to see a diary of world events expressed as a blockchain - forks and all. Is this at least within Redstone's capabilities ? In each world a cave where the veil gr
634.
by deepnet 11y ago
Did they fix it ? Perhaps they cannot fix it ? Maybe their IT is in denial ? Interested to see a follow-up report.
635.
by deepnet 11y ago
This is like punching a guy who hands you the wallet you just dropped. This was entirely RESPONSIBLE DISCLOSURE. They need to send a basket of muffins to the guy. Surely they should take INTENT into account. The interesting question is : Ho
636.
by deepnet 11y ago
If that is the case then why did the evil Robots have no Asimov Laws ? The good butler robots had the 3 Laws. The dreamer hero robot had no laws but had imagination and developed its own morality.
637.
by deepnet 11y ago
In the case of music , nearly all of the money goes to corporations and thus shareholders. It does not go to making new art. DRM and anti-share laws benefit the big players who seek to control culture. They use this money to manipulate the
638.
by deepnet 11y ago
https://www.ted.com/talks/jeremy_howard_the_wonderful_and_te... Jobocalypse
639.
by deepnet 11y ago
In theory yes. In practice the UK poverty trap is very real and crushing to the entrepreneurship of the economically disenfranchised.
640.
by deepnet 11y ago
Yes a great phone but Nokia often gave up on phones soon after release, still a sweet machine potentially capable of repurposing. If a company will not support a device and has locked it down there should be a moral obligation to allow othe
641.
by deepnet 11y ago
Stuart Russel of AI a Modern Approach fame ( with Norvig ) asks us to consider the ethics of AI research - killer robotics, & hunter drones. http://www.nature.com/news/robotics-ethics-of-artificial-int... He calls
642.
by deepnet 11y ago
Many Cell plans send your phone number to any websites you visit.
643.
by deepnet 11y ago
5-7 years 18 months is a joke
644.
by deepnet 11y ago
The latest update 5.02 has been bricking Nexus Tablets, out of warranty. I stopped upgrading because of this. Security and OS updates should be separate
645.
by deepnet 11y ago
Post Snowden this percentage has decreased.
646.
by deepnet 11y ago
I read a lot on a screen. This is tiring. Adverts make it impossible to read the web - they are designed to catch the eye - even when still they are eye catching. To concentrate on text near an advert takes extra mental resources and I
647.
by deepnet 11y ago
Conversely the conscious mind can also make it harder to learn things. I refer to the "Inner Game of Tennis" by Tim Gallwey https://www.youtube.com/watch?v=ieb1lmm9xHk As a teacher he tries to distract the conscio
648.
by deepnet 11y ago
Don't miss C Olah's blog where Figure.1 was copied from http://colah.github.io/ Very very visually insightful on the nature of Neural Nets, Convnets, Deep Nets...
649.
by deepnet 11y ago
Go large. Do some fraction of something like this http://blog.jenniferdewalt.com/post/56319597560/im-learning-... Learn and make a Portfolio and get a lot Publicity at the same time - when you ask for work people
650.
by deepnet 11y ago
Secret Corporate Control Of Many Government that achieves its aims via secret threats and manipulation ? Seems textbook fascism
651.
by deepnet 11y ago
How not ?
652.
by deepnet 11y ago
That is literally the definition of a fascict dictatorship. This is what Adam Smith warned us against. I think you have hit the nail on the head. Governments around the world have capitulated to this. This seems like the financial wing of t
653.
by deepnet 11y ago
In "Learning to Execute" by Zaremba & Sutskever http://arxiv.org/abs/1410.4615 An RNN learns snippets of python Their next paper is "Reinforcement Learning Neural Turing Machines" http:/&
654.
by deepnet 11y ago
Also Learning Contact-Rich Manipulation Skills with Guided Policy Search Levine, Wagener, Abbeel http://rll.berkeley.edu/icra2015gps/robotgps.pdf http://rll.berkeley.edu/icra2015gps/
655.
by deepnet 11y ago
The robot learns to see it's arm and the target using a Imagenet trained CNN. The robot is then shown the task a few times. A human controls the robot for a few minutes, performing the task. Then an innovative policy search finds a rob
656.
by deepnet 11y ago
Learning motor torques directly from vision is a very important result. https://youtu.be/EtMyH_--vnU This talk by Sergey Levine, Pieter Abbeel's PostDoc outlines Berkley's end-to-end deep-training visuomotor contr
657.
by deepnet 11y ago
Learning to do something is an AI problem. Writing a program to play Chess is not AI but doing so has helped figure learning out.
658.
by deepnet 11y ago
You need an error signal - a target value is compared with the networks prediction. That error is carefully assigned proportionally to the network weights that contributed to it and the weights adjusted a small amount in that direction. Thi
659.
by deepnet 11y ago
Great use case, upscaling print thumbnails. Norman Tasfi made a Neural Net upscaler for Flipoard http://engineering.flipboard.com/2015/05/scaling-convnets/ I expect video upscaling next.
660.
by deepnet 11y ago
The FBI are liars and their informants are liars. The real secret the security services are trying to hide is their complete ineffectiveness, total waste of money and hijacking for political and public manipulation. I really don't beli

More ›