Hacker News: chaisan

Autoresearch for Integer Factorization

chaisan — Sun, 29 Mar 2026 18:24:49 +0000

Article URL: https://github.com/iliazintchenko/agent-factoring

Comments URL: https://news.ycombinator.com/item?id=47565725

Points: 4

# Comments: 0

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 16:18:57 +0000

have examples?

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 16:18:47 +0000

nice. for which problem?

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 16:18:07 +0000

yess. loads of space for further exploration here. there is an attempt to keep things as general as possible in the expert.md file, but hard to mitigate overfitting fully. however, changing the seed will not get you much further with all else in the solver constant. unless you try a number of seed that exponentially scales with the size of the problem

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 16:14:52 +0000

sure. in the limit, everything is parameter tuning. with large enough NP-hard problems, the complexity of the search space is big enough that its infeasible to get to a better state by just tuning params in any reasonable amount of time.

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 06:46:59 +0000

wrt. token usage?

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 06:46:37 +0000

somewhat

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 06:46:07 +0000

and it would take an algo change to the solver to jump to the next local optimum

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 06:45:34 +0000

yeh. ofc. but on any problem larger than 40 variables, the gains from random restarts or initializations will quickly plateau

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 04:08:48 +0000

the sum of the weights of the unsatistied clauses. we want to reduce this number

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 04:08:19 +0000

as its from 2024 (MaxSAT was not held in 2025), its quite likely all the solvers are in the training data. so the interesting part here is the instances for which we actually got better costs that what is currently known (in the best-cost.csv) file.

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 04:05:24 +0000

its just comparing the cost of the best solution found to the best known cost we had before. O(N). why optimistic?

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Thu, 19 Mar 2026 03:43:16 +0000

nice. EDA indeed one of the top applications of SAT

Autoresearch for SAT Solvers

chaisan — Thu, 19 Mar 2026 00:40:32 +0000

Article URL: https://github.com/iliazintchenko/agent-sat

Comments URL: https://news.ycombinator.com/item?id=47433265

Points: 167

# Comments: 32

New comment by chaisan in "Autoresearch for SAT Solvers"

chaisan — Mon, 16 Mar 2026 19:06:58 +0000

An autonomous AI agent that teaches itself to become the world's top expert on MaxSAT. Given weighted MaxSAT instances, it learns novel strategies, finds better solutions and iteratively refines its toolbox. No human guidance.