Would companies have to worry about their proprietary code being seen and used in a not-so-obvious way?
If some start up is working on some novel ML algorithms, that has some nice demos out in the public showcasing their work,
I wonder if these bigger companies would take a peek at the source code and use some ideas from the algos for their ml products.
This would prevent start-up from expanding into other areas.
I am new to this space but after reading about your goals for this AI, does this mean that most news sites are just copying facts off one other?
Secondly how does fact checking work in cases like this with AI?
I would imagine fact checking would get progressively more difficult as more articles become AI generated.
My biggest fear is, we will have one AI generating a fake story and and acting as the root node, then other AIs creating news off those this node and we will have fake stories spread in various forms like branches creating misinformation.
> does this mean that most news sites are just copying facts off one other?
This is how AP/Reuters/"newswire" content works. Since very few organisations can afford to have people everywhere, there's a sort of syndication model going on where the people that are in the right place for breaking news can sell it to a broker, who writes it up and then it appears across the rest of the media.
Press releases are even more of a thing. An organisation which wants to get in to the news will make a press release, which is basically a kit of quotes and "facts" that can be loosely re-written into an article.
It's also important to understand how limited journalistic "fact checking" really is. They're not detectives, nor scientists, and don't have the time to be either. Most of the time it consists simply of ringing someone being talked about, asking them "is this true", and printing the result.
1. I'm no journalist, but I'm fairly sure there's some news sites that break the news first and then others follow (either through primary or secondary sources)
2. I haven't implemented fact-checking algos. At the moment I'm not planning to use AI for it, just simple cross referencing (not finalized).
3. When we do start momentum, we'll get on board a highly specialized ethics team and safety team. We're not a state actor or influenced by anyone so we're just creating gimmick-y articles on the internet for now. We'll eventually combat those issues if and when similar services pop up.
One reason we want to be first is that we know state actors are going to pop up soon. Singapore, Russia, Philippines, the list goes on for which they want something like us to control propaganda. We're hoping we'll be able to gain enough momentum before they start (or get further along than they already are at) to set a good solid standard. But with all things, this will take time and deliberation.
Is C/C++ still worth learning if o want to create some models from scratch (new layers or different paradigms)
I hear that C++ is a nightmare to work with and was wondering if Rust,Julia, or even Swift would be worth learning instead.
I know Python but deep learning frameworks seem to be written in C++, so to come up with new layers I need to understand C++, which I was told has lot of peculiarities that takes time to pick up. Compiler isn’t also very user friendly (what I’ve read)
C++ is not as tricky as people make it out to be. There is a lot of elitism among programmers, and a lot of people seem to claim it’s hard solely to make themselves look smarter for being able to write it.
If you know the basics of programming and have the persistence to. RTFM (Read The Fucking Manual), C++ will not give you any trouble. In fact, you might actually start to enjoy it more than the other languages you used in the past.
All that said, if you are focusing on machine learning rather than programming, then you should look into Python and R. A great resource is “an introduction to data science with R” by David Langer: https://m.youtube.com/watch?v=32o0DnuRjfg
If this is the case I would actually love to play around with C++ as a lot of software that Python wraps around is written is in it and it gives me chance to look a little deeper into the source code.
Julia is a blast to do research on this stuff in, if you want to go beyond the basics like TensorFlow and PyTorch allows. The 2020's is going to be the decade of mixing numerical PDEs with machine learning IMO, and Julia already has a lot of features along these lines that are missing from "traditional ML" libraries.
Interesting.
I was going to go through their yearly conference talks to get an sense of Julia’s capabilities.
JuliaCon2019 etc on youtube. Is that the best way?
You can actually implement most new layers or experimental ideas using frameworks like pytorch or tensorflow. They support fairly low-level primitives which are much more flexible than keras or pytorch sequential models.
That said C/C++ is still very useful for implementing high performance systems.
Ah. I haven’t played around with Pytorch custom layers enough so I am going to give it a try. I was initially trying to do it in keras but Keras was just using tensorflow layers for most operations so I couldn’t tweak the original tensorflow layers through keras easily.
The concept of "layers" is not in fact enforced by pytorch or tensorflow at all. This tutorial is a really nice overview of the levels of abstraction available in pytorch
https://pytorch.org/tutorials/beginner/nn_tutorial.html
No point unless you have an interest in numerical linear algebra. The people who write the foundational Fortran/C/C++ libraries are experts in numerical analysis which is another rabbit hole.
If you want to write your own for fun, then there are some great algebra libraries in C++ you can use or you can use bindings for PyTorch or TF.
Yeah I don’t want to write my own libraries but create new layers from the existing numerical algebra layers.
I was originally trying to create a new type of convolution layer in Keras and asked in their official google board, stackoverflow etc , after being stuck for a while but the answers I got weren’t solving the problem.
I haven’t tried creating custom layers in Pytorch yet though so maybe it’s possible to do so with Pytorch and can just learn C++ for other purposes.
The repository owner can file DMCA counterclaim; once they have done so, GitHub will reenable the repository after 2 weeks unless legal action is initiated (not just filing a DMCA).
> We won't refuse to process an otherwise complete counter notice if you don't include this statement; however, we will know that you haven't read these guidelines and may ask you to go back and do so.
