- Oct 2023
-
www.nature.com www.nature.com
-
Another approach for using neural networks to solve mathematical problems is transforming a mathematical formula into a binary sequence of symbols. A neural network policy can then probabilistically and sequentially grow the sequence one binary character at a time6. By designing a reward that measures the ability to refute the conjecture, this approach can find a refutation to a mathematical conjecture without prior knowledge about the mathematical problem.
A nice idea to learn a formula of symbols which can be evaluated logically for truth. But do they mention more general approaches such as using SAT solvers for this task? See Vijay Ganesh work.
-
- Jan 2017
-
en.wikipedia.org en.wikipedia.org
-
Boolean satisfiability problem
This is just one specific type of the classes of satisfiability problems (a.k.a. search problems).
Other related problems include: Linear equation satisfiability, Linear inequality satisfiability, 0-1 integer linear equation satisfiability.
Given the current context (of search problems), all the above are known as NP problems in general (with the observation, that the classic definition of NP limited the scope to only YES-NO problems).
One can think of search problems as "one of many ways of characterizing the set of problems that form the basis of the study of intractability". Other ways include viewing such problems through the lenses of decision problems or optimization problems. In other words, problems in any of the aforementioned types can be translated between (or more formally, reduced to) each other with relative ease.
-