TalkChess.com

Reading myself I found it too hard to understand to not clarify. We do know which of the myriad of possible following states gave us that value. By that I meant the board setting of which the eval() was propagated. After each search we got pairs of states of approximately the same value even though ...

you could double the hash code size Maybe in the age of SSDs that is possible but I find it to be besides the point. In this case it seem to be a implementation problem, not one of too many positions analyzed. I find your entire thinking process / attitude frightening. I hope you do not work at Boe...

mar wrote: ↑

Tue Jan 28, 2020 7:28 am

If his engine crashes, then he should fix his hashmove validation.

I agree. Do not botch/fudge it, fix it.

DustyMonkey wrote: ↑

Mon Jan 27, 2020 4:14 pm

an autoencoder cannot reduce the state much beyond a well packed traditional encoding

Actually it alone can not at all.

Then (as far as I understand, that's what you had in mind) an encoder would only be used to map a discrete board state into a lower dimension real-valued vector that would be easier to put into a kernel/NN/etc. to predict the value, and would also make the training easier. That is right. It is a mo...

an autoencoder cannot reduce the state much beyond a well packed traditional encoding My guesstimate is that it takes ~175 bit but certainly <185 bit to encode any position. That is way too little to encode any meaning with it. How do you intend to represent the board as a cardinal, metric scale, a...

The search term you need is "autoencoder" Yes, thank you. This time I actually knew that. but there will be the issue of what you will want to be considered 'typical' positions - an autoencoder cannot reduce the state much beyond a well packed traditional encoding _unless_ you decide that some 'pos...

PK wrote: ↑

Tue Jan 21, 2020 11:47 am

If one is better

Please specify.
Is that a systematic form of extremizing as in partial information aggregation for Forecasts/Predictions ?

After the search those moves that were investigated the deepest have reliable new evaluations. Thus the available moves now have their original evaluation and a new evaluation identical to that of a much later/deeper move. We do know which of the myriad of possible following states gave us that valu...

I hope my musings might inspire others to look for a 'sweet spot' in the eval/search balance. I am afraid that we can be certain, that you are nowhere near it, from a purely theoretic approach looking at the compute balance of move generation and tracking to evaluation. Think about how much less ti...