Zulip Chat Archive

Stream: Lean Together 2025

Topic: Daniel Weber and Vlad Tsyrklevich: Equational Theories Proj


Riccardo Brasca (Jan 14 2025 at 17:09):

A thread for discussion about this talk.

Xavier Généreux (Jan 14 2025 at 17:21):

Thank you for the talk!
There is mention of the convert tactic being too greedy in the talk.
Did you encounter this problem as well with convert h using 1 or convert h using 2 which caps the number of iterations of the congruence algorithm?

Vlad Tsyrklevich (Jan 14 2025 at 17:22):

The slides are here

Daniel Weber (Jan 14 2025 at 17:27):

Xavier Généreux said:

Thank you for the talk!
There is mention of the convert tactic being too greedy in the talk.
Did you encounter this problem as well with convert h using 1 or convert h using 2 which caps the number of iterations of the congruence algorithm?

When limiting the number of iterations sometimes it doesn't go deep enough

Shreyas Srinivas (Jan 14 2025 at 17:30):

Just a small correction. The project management stuff was designed by both me and Pietro

Vlad Tsyrklevich (Jan 14 2025 at 17:31):

Oops, apologies for getting that wrong!

Shreyas Srinivas (Jan 14 2025 at 17:31):

This is roughly where it began: #Equational > PRs Welcome : CI actions to manage the project @ 💬

Shreyas Srinivas (Jan 14 2025 at 17:32):

basically I set up the github project and designed the FSM and Pietro did the heavy lifting by writing the CI scripts to implement the FSM*. This was followed by several iterations of improvement.

*Finite State Machine

Pietro Monticone (Jan 14 2025 at 22:49):

:video_camera: Video recording: https://youtu.be/v_qL3Z_l47s?si=kKpBwtwkuJXDG44M

Siddhartha Gadgil (Jan 15 2025 at 06:06):

Thanks. I really enjoyed the talk.

I was wondering how deep is the integration of TPTP with Lean? Is there a type/typeclass/DSL in Lean for the TPTP format? Is there some mechanism to translate problems to TPTP if they are of the correct form?

Vlad Tsyrklevich (Jan 15 2025 at 07:21):

In Lean/Mathlib itself, there is no TPTP support as far as I'm aware. The Equational Theories Project translated from/to TPTP; however, I've also used Duper a bit and Duper does have some support for TPTP in Lean, like this. I'm not very familiar with what's there, so I'm not sure if you can translate _to_ TPTP, but clearly you can do some translation from TPTP as that example shows.

Terence Tao (Jan 15 2025 at 08:38):

Great talk and slides! I think many points from the slides could be incorporated into the paper (if they are not already there).


Last updated: May 02 2025 at 03:31 UTC