Guillaume Dalle

gdalle.bsky.social

01 July 2025 09:47

Just discovered an awesome paper on inserting linear programs into neural networks: arxiv.org/abs/2505.22224 Most existing methods require solver calls during traning, which is very expensive. This work introduces a new (differentiable, convex) loss which bypasses the solver using geometry!

Samuel Vaiter

samuelvaiter.com

30 April 2025 08:12

The SAT problem asks whether a logical formula, composed of variables and (AND, OR, NOT), can be satisfied by assigning True to the variables. SAT is NP-complete along with 3-SAT, with clauses of three literals, while 2-SAT, is in P! buff.ly/mMLGbw4

Samuel Vaiter

samuelvaiter.com

11 April 2025 13:06

Opinion of the day: we don't desk reject enough in ML. Too much energy is wasted in 4x reviewing papers that will *obviously* be rejected. Second opinion otd: we don't teach enough students to be positive. We should not seek how to reject a paper, but how to accept it. And yes, #1 has a role in #2

Motonobu Kanagawa

motonobu-kanagawa.bsky.social

17 March 2025 08:16

Donkeys absorbing sunshine...

Han Bao

han-b.bsky.social

20 February 2025 19:13

Today I met two guys each from France and Burkina Faso, tried my poor French and got “pas mal” “presque parfait”, which is not satisfactory enough but already a good achievement for me now😂