TalkChess.com

pohl4711 wrote: ↑Sun May 27, 2018 10:31 am

mjlef wrote: ↑Sun May 27, 2018 2:21 am You give no evidence that a neural network is needed for Monte Carlo Tree Search

I know. Because I never said something like this. All I say is, that the interesting, risky positional of LCZero comes from its neural network, not from the MCTS, which LCZero uses only, because the neural-net calculations are so slow, that using normal AlphaBeta with crunching billions of nodes, would not work properly.
I looked into some dozend games of Komodo MCTS and find it playing like Komodo: good, solid,.positional chess, but much weaker than Komodo 12, because no real AlphaBeta search is done (seems at one level to Wasp 3). And that is, what is to be expected, when MCTS is used without a neural net.
So, Komodo MCTS has nothing in common with LCZero, but the using of MCTS, which is not only useless without a neural net, but weakening. And so, I see no reason to use Komodo MCTS.
Thats all I said. And that is all I have to say about Komodo MCTS, until it uses a neural net.

And because, I believe, that the Komodoteam knows all this like me (better like me), I think, the whole MCTS-Komodo was done to suggest parallels to AlphaZero/LC Zero, which are not existing. And I dont like that.
That has nothing to do with Komodo itself. I like Komodos positional play and its solid evaluations very much and I used only Komodo for all my work on my SALC-openings. I believe, there is no better engine for analyzing, than Komodo. Now and since the last years.

That seems like an unfair accusation to me. MCTS is nothing new you might remember that it was an analysis feature in Rybka 3:
https://en.chessbase.com/post/rybka-s-m ... o-analysis

From my understanding MCTS is just a method of analysis in which many games are played out from the chosen position to see which move is more likely to give you the desired outcome. Neural networks are an entirely different thing.

Leto wrote: ↑Sun May 27, 2018 2:48 pm

pohl4711 wrote: ↑Sun May 27, 2018 10:31 am

mjlef wrote: ↑Sun May 27, 2018 2:21 am You give no evidence that a neural network is needed for Monte Carlo Tree Search

I know. Because I never said something like this. All I say is, that the interesting, risky positional of LCZero comes from its neural network, not from the MCTS, which LCZero uses only, because the neural-net calculations are so slow, that using normal AlphaBeta with crunching billions of nodes, would not work properly.
I looked into some dozend games of Komodo MCTS and find it playing like Komodo: good, solid,.positional chess, but much weaker than Komodo 12, because no real AlphaBeta search is done (seems at one level to Wasp 3). And that is, what is to be expected, when MCTS is used without a neural net.
So, Komodo MCTS has nothing in common with LCZero, but the using of MCTS, which is not only useless without a neural net, but weakening. And so, I see no reason to use Komodo MCTS.
Thats all I said. And that is all I have to say about Komodo MCTS, until it uses a neural net.

And because, I believe, that the Komodoteam knows all this like me (better like me), I think, the whole MCTS-Komodo was done to suggest parallels to AlphaZero/LC Zero, which are not existing. And I dont like that.
That has nothing to do with Komodo itself. I like Komodos positional play and its solid evaluations very much and I used only Komodo for all my work on my SALC-openings. I believe, there is no better engine for analyzing, than Komodo. Now and since the last years.

That seems like an unfair accusation to me. MCTS is nothing new you might remember that it was an analysis feature in Rybka 3:
https://en.chessbase.com/post/rybka-s-m ... o-analysis

From my understanding MCTS is just a method of analysis in which many games are played out from the chosen position to see which move is more likely to give you the desired outcome. Neural networks are an entirely different thing.

MCTS has no rollouts, and has nothing to do with Monte Carlo simulations.

lkaufman wrote: ↑Mon May 14, 2018 8:11 am Currently it is not able to benefit from running on more than three threads, but we expect to raise or eliminate this limit in the next month or so and if so we will offer one update (together with other improvements to the MCTS version) free to all purchasers of Komodo 12.
...
4. Eventually it should be able to benefit from many cores more than normal Komodo, although that is obviously not true yet.

Think so too, Larry.
But the kombination of k12N(ormal) and M(CTS) yet seems of practical usefulness.
Until the two settings could use a shared hash once yet too, I do save the hash of the one of them and reload it with the other one.
Works well.
More cores for MCTS would help at least to fill hash more quickly.

pohl4711 wrote: ↑Sun May 27, 2018 10:31 am

mjlef wrote: ↑Sun May 27, 2018 2:21 am You give no evidence that a neural network is needed for Monte Carlo Tree Search

I know. Because I never said something like this. All I say is, that the interesting, risky positional of LCZero comes from its neural network, not from the MCTS, which LCZero uses only, because the neural-net calculations are so slow, that using normal AlphaBeta with crunching billions of nodes, would not work properly.
I looked into some dozend games of Komodo MCTS and find it playing like Komodo: good, solid,.positional chess, but much weaker than Komodo 12, because no real AlphaBeta search is done (seems at one level to Wasp 3). And that is, what is to be expected, when MCTS is used without a neural net.
So, Komodo MCTS has nothing in common with LCZero, but the using of MCTS, which is not only useless without a neural net, but weakening. And so, I see no reason to use Komodo MCTS.
Thats all I said. And that is all I have to say about Komodo MCTS, until it uses a neural net.

And because, I believe, that the Komodoteam knows all this like me (better like me), I think, the whole MCTS-Komodo was done to suggest parallels to AlphaZero/LC Zero, which are not existing. And I dont like that.
That has nothing to do with Komodo itself. I like Komodos positional play and its solid evaluations very much and I used only Komodo for all my work on my SALC-openings. I believe, there is no better engine for analyzing, than Komodo. Now and since the last years.

Monte Carlo Tree search is pretty well described here: https://en.wikipedia.org/wiki/Monte_Carlo_tree_search

There are many variations on it. The scheme estimates win probabilities via a number of means. A classic approach is when hitting some leve in the search tree, ply play a game with random moves from that position (called playout or rollouts), then back the result (in chess, win, loss or draw) up the tree. Each tree node contains a sum of the win/loss/draw and a count of the number of times visited. The winning percentage at each point in the tree guides it search either by exploiting what you know (favoring the best scoring move) plus exploring (favoring less visited nodes). Programs like Alpha Zero use a neural network to give win probabilities at a new node, instead of the playouts.rollouts. But it is still called Monte Carlo Tree Search since the exploitation/exploration and summing win probabilities remains the same. I think the ALpha Zero/Leela neural network is fascinating and I am anxious to see how well it can learn. But that neural network is not needed for Monte Carlo. In Komodo, we do not currently use a neural network in MCTS mode. We use a heavily modified evaluation and combined with short searches to in place of the neural network.

Larry and I have discussed this approach for years. The great performance of Alpha Zero did push us to work on it, although last tear even before Alpha Zero came out, I had talked with Larry about starting work on MCTS this year.

I do not know if Komodo's eval and search will be better or worse than Alpha Zero/Leela's neural network. I hope it will be but we have a few hundred more elo to go. But the progress since the K 12 release has been really good. We like the way it plays. And we find it coming up with very interesting moves.

I would love someone to try another experiment. Take the Leela neural network, and just write a classic search for it, using the Leela neural network in place of an eval, but include all the standard cutoffs. If easier, you could convert the win percentage back to a centipawn score, and try all the standard pruning tricks we do in programs like Stockfish. I would love to see how it performs. Neural networks can in many ways see future moves via piece cooperation, attack and defense terms. It might be a very powerful program.

Hurnavich wrote: ↑Sun May 27, 2018 6:26 am Hi,
I'm still curious as to why MCTS is not working in fritz GUI
The GUI states that the engine is not supported
If the development of this MCTS engine improves
Then at some point I would like it to run in fritz
And on my system it does not
Yet all normal Komodo does

Thanks

I have no problem with Komodo MCTS in Fritz 16. It works fine.

I get it running but it has slow speed and depth and not sure what it is supposed to be doing

K12 normal works perfect

Hurnavich wrote: ↑Mon May 28, 2018 3:03 pm not sure what it is supposed to be doing

Good question, but no good answer.
How about "waiting for NNs "

Just my opinion.

I have been doing some experimenting on K12 mcts and Stockfish eval given a LTC. I have to generally agree with Mark and Larry that K12 comes up with interesting novelties that SF will prune out. I also find that K12 MCTS plays very dynamic chess that is more akin to human play. When positions are analyzed further down the tree both Sf and K12 evals trend toward each other. I of course do not have enough imperial evidence(very small sample size) for this theory.

Ok we must know is this Komodo strong or no. How much time new Komodo MonteCarlo need to find moves from famous game AlphaZero vs Stockfish (game with queen on h-line)?

mjlef wrote: ↑Mon May 28, 2018 4:22 am I would love someone to try another experiment. Take the Leela neural network, and just write a classic search for it, using the Leela neural network in place of an eval, but include all the standard cutoffs. If easier, you could convert the win percentage back to a centipawn score, and try all the standard pruning tricks we do in programs like Stockfish. I would love to see how it performs. Neural networks can in many ways see future moves via piece cooperation, attack and defense terms. It might be a very powerful program.

Interesting idea. I thought about this before