@johnregehr I just watched your 2014 talk on Souper at UW on YouTube. Fascinating how similar it is (was at the time) to "freduce" in Yosys.
-
-
Replying to @johnregehr
Fig. 2 looks exactly like the procedure I hinted at. (S is set of in-out samples.) I'll read the whole thing tomorrow.. Thanks for the ptr!
1 reply 0 retweets 1 like -
Replying to @oe1cxw @johnregehr
Consider this: A circuit with 32 inputs and 32 outputs but no operators, just direct connections between in- and output bits.
1 reply 0 retweets 0 likes -
Replying to @oe1cxw @johnregehr
That's already a search space of 32^32 configurations (there is no requirement to use all inputs). And you'd need at least 32 samples to
1 reply 0 retweets 0 likes -
Replying to @oe1cxw @johnregehr
uniquely constrain to one solution. Of course that algorithm would still work fine for this circuit. This is just to illustrate the search
1 reply 0 retweets 0 likes -
Replying to @oe1cxw @johnregehr
space explosion that comes with bit level wiring. Throw actual operators in the mix and the number of samples you need will explode.
1 reply 0 retweets 0 likes
I know Alan Mishchenko experimented with something like this for very small boolean circuits. But I don't think he published yet.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.