David Quarel
AboutNowContact

SLT for RL animations

Sep 10, 2025

A place for me to dump links for work on SLT for RL.

  • LLC for open maze, cheese always in corner
    • hf_checkpoints
    • animation
  • LLC for open maze, cheese can be anywhere
    • hf_checkpoints
    • animation

Extended training runs:

  • Open maze, cheese always in corner. Checkpoint every 50 batches.
    • discount \(\gamma = 0.99\)
      • hf_checkpoints
      • animation
    • discount \(\gamma = 1\)
      • hf_checkpoints
      • animation

Subscribe

  • David Quarel
  • davidquarel[at]gmail[dot]com

Last Update:

2025-09-10

19:53:47 UTC