Yutaka Ng
Yutaka Ng
Dear @DyeKuu , I guess nobody has the admin right to handle pull requests to this repository and the development of miniF2F is continuing at [the corresponding repository at Facebook...
- Domain-specific optimisations? - [ ] Avoid template-based conjecturing when no user-defined types appear in the proof goal. - [ ] More tactics for math problems. - [ ] math_olympic...
This would be useful for debugging as well.
1. We apply the `induct` tactic. 2. We get a step case whose shape looks like `induction_hypothesis ==> conclusion`. 3. Take `conclusion` from this step case. 4. Apply `clarsimp to...
- bottom-up conjecturing or goal-oriented conjecturing? Both: bidirectional conjecturing. - structural proofs or backwards proofs? - Both: structural proofs for bottom-up conjecturing and backwards proofs for goal-oriented conjecturing. - Note...
We should be able to reason about types of HOL to efficiently produce types.
For example, if two alpha-equivalent nodes appear twice in the same path, probably something has gone wrong. We can improve the accuracy of pruning by developing assertions on the log/path.
Can we get rid of chained states from `seed_of_or2and_edge`? Instead of passing a chained state that we obtain after applying certain proof methods, I would like to reconstruct such proof...