vonnik

106 ms·

931.
by vonnik 11y ago
It seems like a no-brainer that one of the largest programming languages in the world should have strong machine learning tools. Maybe the reason ML isn't done in Java and Scala is simply because the tools haven't been there. Pyth
932.
by vonnik 11y ago
This is fantastic. For deep learning in Java or Scala to feed feature vectors into Xyclade, we built http://deeplearning4j.org
933.
by vonnik 11y ago
Yoav is one of the smartest guys in this space. He and Omer Levy also wrote a great explanation of word2vec: http://arxiv.org/pdf/1402.3722v1.pdf https://levyomer.wordpress.com/2014/04/25&#x2F
934.
by vonnik 11y ago
They have set out to aestheticize a series of projects where humans do not live well. We consume these photos as art, but the feeling of profound alienation and danger inherent in such ill-formed spaces doesn't lead to healthy communit
935.
Google voice search: faster and more accurate(googleresearch.blogspot.com)
10 points by vonnik 11y ago|0 comments
936.
by vonnik 11y ago
Great piece. Our startup supports http://deeplearning4j.org and http://nd4j.org , and it's a great feeling having the open-source wind at your back. Talented strangers helping to fix and grow your project. To use
937.
DeepMind: Deep Reinforcement Learning with Double Q Learning(arxiv.org)
3 points by vonnik 11y ago|0 comments
938.
by vonnik 11y ago
This article, published in May, is already somewhat dated, and demonstrates some fundamental misunderstandings about the market. Human wealth advisors and robo-advisors only overlap for a limited number of services, notably asset allocation
939.
by vonnik 11y ago
Yeah, I think there's more acknowledgement in Europe of the sacrifices the USSR made during the war, particularly in countries like France with a strong communist tradition. But winning the war and reconquering France were two differen
940.
by vonnik 11y ago
Signing an Armistice is not a victory. The French and Germans both signed, after French soil was occupied for years and many of its best farmland destroyed during the war. The Germans were outmaneuvered during the negotiations of the Treaty
941.
by vonnik 11y ago
Germany may have been defeated at the end of WWII, but France was routed during the war. The Germans occupied the northern parts of the country, and were widely accepted by a population of collaborators and anti-Semites. The Vichy governmen
942.
by vonnik 11y ago
Generalizations about entire nations, and about the French in particular, are tiresome and usually trite. The French are incredibly successful at producing very smart people, whom they have exported to other economies for the last several d
943.
Clojure: Core.matrix protocols with ND4J(github.com)
2 points by vonnik 11y ago|0 comments
944.
by vonnik 11y ago
You can find out more here if you have questions: https://gitter.im/3tl/usptoio
945.
Show HN: Patent Time Machine – Who Owned What When(uspto.io)
7 points by vonnik 11y ago|6 comments
946.
by vonnik 11y ago
Fwiw, a deep-learning startup that's doing some pretty cool satellite imagery analysis is Orbital Insight http://orbitalinsight.com/ . So deep nets are all over deep space... (I'm aware of them because I work on an
947.
by vonnik 11y ago
I'm part of an open-source project commercially supporting two main libraries, a deep-learning framework, Deeplearning4j -- http://deeplearning4j.org -- and a scientific computing library, ND4J (Numpy for Java), to power it
948.
by vonnik 11y ago
I'd like to offer a couple more in-depth approaches: * Use Pimsleur tapes, if you can, to master the basic conversational phrases. Downloadable all over the place. * Apps like Memrise use a similar delayed repetition method. Read up on
949.
Who Owned Which Patents When(uspto.io)
2 points by vonnik 11y ago|0 comments
950.
by vonnik 11y ago
This is a really neat, rules-based, Chomskyian NLP (as opposed to the statistical kind represented by Word2vec). It's an old division... http://norvig.com/chomsky.html The essential question is: Can it go from basic re
951.
by vonnik 11y ago
It will take one hundred years or another world war to clear out all the crap that France makes businesses and individuals go through to live there and make a living. Hollande's Socialists, while well meaning, are fairly incompetent co
952.
by vonnik 11y ago
Not sure that the premise of this article is even correct. I can think of a lot of successful European startups -- Minecraft, Spotify, Skype, DeepMind, the list goes on... -- and just as many startups led by Europeans that have succeeded in
953.
Deeplearning4j on Spark(deeplearning4j.org)
3 points by vonnik 11y ago|0 comments
954.
by vonnik 11y ago
I think there is value in what you are doing, as long as everyone knows what context to put the information in, and realizes that they may lack significant additional information. In the wild, HFs are actively engaged in sending misinformat
955.
by vonnik 11y ago
This is a neat idea, but it has a flaw. The flaw is that Navlio is basing the recommendations on long equity positions (as well as other securities reported in HFs' 13F filings). But hedge funds hedge their bets -- hence the name -- by
956.
DeepMind: Continuous control with deep reinforcement learning(arxiv.org)
1 points by vonnik 11y ago|0 comments
957.
by vonnik 11y ago
Great article. I've recruited engineers and worked with much more experienced recruiters. Before LinkedIn, they used to "smile and dial", which was just going down a list of numbers at a company trying to get names and positi
958.
by vonnik 11y ago
Sebastian Thrun teaches a course called AI for Robotics on Udacity: https://www.udacity.com/course/artificial-intelligence-for-r...
959.
by vonnik 11y ago
I agree. Andrew Ng's course is a great place to start with ML. We've got a list of great online ML/DL courses here: http://deeplearning4j.org/deeplearningpapers.html
960.
A Beginner’s Guide to Recurrent Networks and LSTMs(deeplearning4j.org)
211 points by vonnik 11y ago|10 comments

More ›