Alphazero news

[matthewlai]

It's hard to say and I don't want to speculate much beyond what we have data to support, but my guess (and I can very well be wrong) is that there's much less diversity when SF uses the BF opening book.

That is correct. When playing with the default settings (one of the default options is BestBookMove=true), BrainFish plays always the best move out of the Cerebellum-Library, which means: BrainFish always plays the same move in the same position.

Regards - Stefan (SPCC (main test-website of Brainfish)) https://www.sp-cc.de/

Yes, but didn't they take SF8 + book itself, and not the full BrainFish with its UCI option? That would be a good match: A0 versus full BrainFish of early 2018 with varied openings UCI option, but they seem to not have done that.

We used the BrainFish player for the book moves, and switch to SF8 (for consistency with other results) once out of book. We did that to make sure we were using the opening book correctly, and we weren't aware of the BF UCI option for diversity, so we only tried enforcing diversity from the AZ side.

[Milos]

Is it because people who wrote those 'fully decent' papers are not Google-branded?

But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

You have no idea how is to publish in "Nature" and "Science" (Deep Mind publishing in both on this same theme). HGM has a "Nature" paper. I don't have, although I do have (few) papers in the best journals of physics (Phys. Rev. Letters, for example). I don't know about morning newspapers, not exactly my field of expertise. Many, many very solid researchers never in their life get to have a "Nature" or a "Science" paper.

So you say, author's institution or group doesn't play any role in paper acceptance chances? Bunch of my colleagues have Nature publications and none of them had any when they were working at university or for smaller companies.

[noobpwnftw]

Is it because people who wrote those 'fully decent' papers are not Google-branded?

But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

You have no idea how is to publish in "Nature" and "Science" (Deep Mind publishing in both on this same theme). HGM has a "Nature" paper. I don't have, although I do have (few) papers in the best journals of physics (Phys. Rev. Letters, for example). I don't know about morning newspapers, not exactly my field of expertise. Many, many very solid researchers never in their life get to have a "Nature" or a "Science" paper.

This is like trying to argue that Hitler must be a good person because he was 'Man of the Year' on TIME.

[hgm]

Is it because people who wrote those 'fully decent' papers are not Google-branded?

But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

Conspiracy theories are the hallmark of paranoia. You have to 'pass' 3 anonymous peer reviewers, and if only a single one of those is slightly critical about the quality or the importance of the result, your paper will be rejected outright. Not every scientist is necessarily a fan of big commercial companies; most are low-payed academics.

Of course it is true that big institutions or prestigious universities often fund more important (usually more expensive) reseach, which by its intrinsic merits has a larger chance to be of interest for a prestigious journal.

[Laskos]

Is it because people who wrote those 'fully decent' papers are not Google-branded?

But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

You have no idea how is to publish in "Nature" and "Science" (Deep Mind publishing in both on this same theme). HGM has a "Nature" paper. I don't have, although I do have (few) papers in the best journals of physics (Phys. Rev. Letters, for example). I don't know about morning newspapers, not exactly my field of expertise. Many, many very solid researchers never in their life get to have a "Nature" or a "Science" paper.

So you say, author's institution or group doesn't play any role in paper acceptance chances? Bunch of my colleagues have Nature publications and none of them had any when they were working at university or for smaller companies.

Sure Deep Mind, as a large, well funded group, has it easier. My younger sister at CERN had several "Nature" papers, with groups of 50 to 400 or so. But anyway, a paper in "Nature" is not "decent", it usually is of significant "decency" for science.

[Milos]

Is it because people who wrote those 'fully decent' papers are not Google-branded?

But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

Conspiracy theories are the hallmark of paranoia. You have to 'pass' 3 anonymous peer reviewers, and if only a single one of those is slightly critical about the quality or the importance of the result, your paper will be rejected outright. Not every scientist is necessarily a fan of big commercial companies; most are low-payed academics.

BS. You obviously never worked in a multi billion dollar company research department. You never get outright rejection, never, not for Science not for Nature, not for any other journal. At worst you get recommendation to resubmit with major modifications. And then you haggle with reviewers, like on a flee market, and the less outstanding your paper is, the longer the haggling lasts. Sometimes, ofc you don't get publication, but if it is important enough for the company to show the results you will eventually push it through.
And regarding anonymous peer reviewing, it is never double-blind even when journal policy says so. Reviewers always know who you are, just by reading the abstract and references list. And after first reviewer's report comes in you also usually know who are the reviewers or can guess with high degree of certainty.
Conferences are even worse, just look at NIPS or E(I)CCV, acceptance rate is like 20% and if you are from Google acceptance rate goes to 70-80%.

[jp]

Is it because people who wrote those 'fully decent' papers are not Google-branded?
But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

You have no idea how is to publish in "Nature" and "Science" (Deep Mind publishing in both on this same theme). HGM has a "Nature" paper. I don't have, although I do have (few) papers in the best journals of physics (Phys. Rev. Letters, for example). I don't know about morning newspapers, not exactly my field of expertise. Many, many very solid researchers never in their life get to have a "Nature" or a "Science" paper.

That's because Nature & Science are for trendy stuff that the public can get excited over. That does not mean trendy stuff the public gets excited over has to be bad, just that work that's extremely good but not trendy and understandable by the public will not be in there.

Conspiracy theories are the hallmark of paranoia. You have to 'pass' 3 anonymous peer reviewers, and if only a single one of those is slightly critical about the quality or the importance of the result, your paper will be rejected outright. Not every scientist is necessarily a fan of big commercial companies; most are low-payed academics.

Of course it is true that big institutions or prestigious universities often fund more important (usually more expensive) reseach, which by its intrinsic merits has a larger chance to be of interest for a prestigious journal.

That is not true. It's definitely possible to get the paper accepted without reviewers all being happy.


But the publication policies of Nature & Science probably shouldn't be the main topic here.

[matthewlai]

Conferences are even worse, just look at NIPS or E(I)CCV, acceptance rate is like 20% and if you are from Google acceptance rate goes to 70-80%.

And they submit papers of equal quality? I don't know about Google, but we only submit paper that we are pretty sure will be accepted. We don't submit the other papers. Anyone can submit papers to NIPS. Most PhD students don't have as high a bar before they decide to submit, and most of them don't have the experience to be able to predict whether some result is significant enough to be accepted.

I am not sure if you really don't understand this kind of basic statistics, or are you just feigning ignorance for as long as it helps your story.

[noobpwnftw]

Isn't it interesting that people just can't help to bring branding tags into discussion.

Now I'm like 90% sure how that went through.

[Milos]

Is it because people who wrote those 'fully decent' papers are not Google-branded?
But in my view a 'decent paper' is still a decent paper even if it is published on a morning newspaper.

You have no idea how is to publish in "Nature" and "Science" (Deep Mind publishing in both on this same theme). HGM has a "Nature" paper. I don't have, although I do have (few) papers in the best journals of physics (Phys. Rev. Letters, for example). I don't know about morning newspapers, not exactly my field of expertise. Many, many very solid researchers never in their life get to have a "Nature" or a "Science" paper.

That's because Nature & Science are for trendy stuff that the public can get excited over. That does not mean trendy stuff the public gets excited over has to be bad, just that work that's extremely good but not trendy and understandable by the public will not be in there.

Exactly and that is why these publication are used as virtual press releases by the big companies.
The procedure is the following, you publish an article in Nature, Science, then you issue an actual press release, then you invite scientific journalists to an open day or similar, then you get a lot of press and finally you announce a new product/technology.
Most ppl are naive or don't know, while others pretend. I'm pretty sure I know it, you know it, Kai knows it, Bojun knows it, Matthew knows it (maybe he doesn't he's just too young ), even HGM knows it but just plays dumb.