From: François Fleuret Date: Sun, 23 Jun 2024 06:52:05 +0000 (+0200) Subject: Update. X-Git-Url: https://fleuret.org/cgi-bin/gitweb/gitweb.cgi?a=commitdiff_plain;h=cdf7c042612ee3b7e5cc2277e82399693a3bf343;p=culture.git Update. --- diff --git a/README.txt b/README.txt index 1bd8dbc..d5cf6d8 100644 --- a/README.txt +++ b/README.txt @@ -1,6 +1,9 @@ Trying to make GPTs build their own "culture". +Francois Fleuret +Jun 21st, 2024 + * Motivation The original motivation of this experiment is the hypothesis that @@ -21,13 +24,13 @@ be solved but not by everybody. There are 5 competing GPTs. The "world" is a 7x9 grid with three "birds" moving in a straight line -and bouncing on the world's borders. The colors correspond to a fixed -"z-buffer order". It could be another "world", but this one has -objectness, occlusion, and motion. +and bouncing on the world's borders. It could be another "world", but +this one has objectness and motion. Given a random world state, and the state after two iterations of birds moving, a "quiz" is to predict the second frame, given the -first, or the opposite. +first, or the opposite. The starting and ending states are chosen, by +rejection, so that there is no occlusion. My home-baked GPT-37M trained with 250k solves this with ~99% success [to be verified with the new setup]. diff --git a/world.py b/world.py index 0d3509f..839f4ff 100755 --- a/world.py +++ b/world.py @@ -153,7 +153,7 @@ def generate_seq( ###################################################################### -def generate_seq_( +def generate_seq_old( nb, height, width,