TalkChess.com

I also doubt that reversibility of the translation process is a requirement for the inheritance of the copyrights from the original. It is also next to impossible to retreive the original source code from a highly optimized binary. You can run it through a de-compiler, but what you would get out would be so different from the original source code that, had someone written it independently, would almost certainly not have been considered a copy-right infringement. The 'perceptible aesthetic' features of the source code (such as the variable names!) would certainly be lost. But I suppose we all agree that stripping the symbol table would not be enough to make a binary free of copyrights.

The output of yacc would still be covered by the same copyrights as the input fed to it. Which is just a program specifying the compiler to be built in a higher-level language. (They do not call it a compiler compiler for nothing...) Possibly in addition with the copyrights on added standard stuff, if not explicitly waived.

Neural nets are data-flow computers, and the network weights are a binary representation of the program they run. That I use a computer program as an aid to designing it should also not matter. Neither being a program, nor being a generated with the aid of a program exclude copyrightability. It all depends on whether their went sufficient human creativity into the process of generating it. It doesn't even have to be relevant creativity, as the example of variable names shows...

chrisw wrote: ↑Tue Feb 23, 2021 1:47 pm ..
But, and I didn’t see it mentioned here as yet, neural network data can’t be back-converted to anything, code or otherwise, created by human hand. It’s a one-way street, irreversible, and the data <==> code equivalence on which the copyright argument depends does not exist. No copyright on NN data.
..

I like this one, what does an NN actually encode, if we assume that NNs are an
approximation of perfect knowledge, can you copyright the weights for perfect
chess knowledge? Regardless if the path to achieve that NN can be formulated
as creative process in a kind of algorithm.

--
Srdja

Copyright law is complex, and us non-lawyers can have opinions about it, but what is copyrightable is not always clear-cut, doesn't always follow what seems intuitive, and can vary by jurisdiction.

The general claim that NN weights are not copyrightable seems untenable. Such weights could in principle be hand-crafted, the result of a creative design process, written as source code and compiled into a binary file of weights. Any network of boolean logic can be emulated by a neural net. Conventional hand-crafted evaluations are in fact neural very simple neural networks. The binary resulting from the compilation of their source code carries copyrights.

syzygy wrote: ↑Tue Feb 23, 2021 1:41 pm But, as you say, the goal is to improve strength. Innovations in Formula 1 cars aren't copyrightable if they aim to increase the car's performance rather than its aesthetics.

You say tomato, I say ...

For many, strong chess play is beautiful.

Separately, does aesthetics have anything to do with "art" these days?

It is the creativity required for top-performing nets that I am highlighting, regardless of whether or not that makes them copyrightable.

Case: ING, a Dutch company, commissioned a "new Rembrandt" to be generated by AI and printed on a 3D printer. There hasn't been a legal test case, but the article below argues that the copyright for this new painting probably belongs to ING:

https://alj.artrepreneur.com/the-next-r ... rated-art/

On this basis, assuming that Chessbase commissioned Mr Silver to produce the FF2 NN, and further assuming that a court would view this NN as an original work (hmm........), then the copyright for that NN would belong to Chessbase.

hgm wrote: ↑Tue Feb 23, 2021 5:16 pm The general claim that NN weights are not copyrightable seems untenable. Such weights could in principle be hand-crafted, the result of a creative design process, written as source code and compiled into a binary file of weights. Any network of boolean logic can be emulated by a neural net. Conventional hand-crafted evaluations are in fact neural very simple neural networks. The binary resulting from the compilation of their source code carries copyrights.

Nonsense. NN weights are streams of numbers and streams of numbers (pi, e and so on) are not copyrightable unless they are demonstrably special in some way. Simply claiming copyright is insufficient, you would have to show that the number stream represented something that was created by human hand. A song, a book, some computer code, a database, whatever. Obviously you can’t put the number stream directly along side the “original human created work”, so, ultimately, you would need to perform AFC test, abstract out of the number stream something that could be put side by side with whatever is claimed as the human created copyrighted thing and then claim a copy-match.
But the NN can’t be reversed backwards, so you can’t do AFC, so any copyright claim is dead in the water.

jdart wrote: ↑Tue Feb 23, 2021 4:30 pm Copyright law is complex, and us non-lawyers can have opinions about it, but what is copyrightable is not always clear-cut, doesn't always follow what seems intuitive, and can vary by jurisdiction.

Exactly. As mentioned at the very start of this thread, I have integrated licensed pretrained models into a commercial application on behalf of of a client. The client’s IP lawyers gave a typical lawyerly opinion on whether pretrained models could be copyrighted and licensed: probably, but it has yet to be tested in court.

In the meantime, here are a sampling of projects and companies claiming licenses (which generally presumes copyright) on their pretrained models:
- pytorch project
- ImageNet (to the point where models trained on ImageNet would be covered)
- Facebook
- Google

There is also an AWS marketplace for pretrained models as a service. (https://aws.amazon.com/marketplace/solu ... ned-models)

Last, pretrained models have been in use in medical imaging solutions for years and are generally covered under the commercial license of the product.

My money is on the courts and Congress in the US siding with commercial interests.

chrisw wrote: ↑Tue Feb 23, 2021 6:12 pmNonsense. NN weights are streams of numbers and streams of numbers (pi, e and so on) are not copyrightable unless they are demonstrably special in some way. Simply claiming copyright is insufficient, you would have to show that the number stream represented something that was created by human hand. A song, a book, some computer code, a database, whatever. Obviously you can’t put the number stream directly along side the “original human created work”, so, ultimately, you would need to perform AFC test, abstract out of the number stream something that could be put side by side with whatever is claimed as the human created copyrighted thing and then claim a copy-match.
But the NN can’t be reversed backwards, so you can’t do AFC, so any copyright claim is dead in the water.

The AFC test (link) could readily be applied to a net. One of the following two conditions would do:

1. the weights files being compared were very similar

2. the two programs gave very similar evaluations (without branching) in a wide variety of different positions (this wouldn't work if the evaluations were actually 100% chess-accurate)

jjoshua2 wrote: ↑Mon Feb 22, 2021 12:21 am

Ferdy wrote: ↑Sun Feb 21, 2021 8:40 pm the protection with regards to ai there may not be underestimated.

Interesting. Just the act of monitoring a process and stopping when the results are artistically pleasing is probably enough, while a pure supervised learning would not be. I am confused why they say a reinforcement learning also counts. Presumably there is a human directing some of the reinforcement direction or at least monitoring hyperparmeters and changing them and determining when to stop the process? That could apply to supervised learning too, so not sure what the distinction they are making is.

It is the unsupervised learning that has the low or zero probability of being granted a copyright. Reinforcement learning in autonomous cars is not easy, the ai systems, NN weights, etc. there has a high probability of getting a protection.

Things may favors NN in the long run, as it is also about the cost of producing the NN weights that also matters. People, Corporations have invested a lot on ai, what is the point of this investment if the NN weight is not protected. The common phrase "only works from human being can be protected under Copyright Act." probably needs reexamination. Programmers can make ai system automate many things a lot more if there are more resources. Are we going to hold back the ability of programmers and not fully use the resources available by letting the human do some works instead of ai so that our NN weights may get a high chance of getting a protection?