Training log: Robot model turns into a 'drunk duck' after step 500

tristanbob · x · 2026-08-30

A developer shared a training log on WandB showing a robot model climbing quickly until turn 85, followed by a slow incline. By model step 500, the performance resembled a 'drunk duck'.

Related event: RL-Trained Virtual Duck Fails to Walk Reliably After 5,000 Iterations(6 posts)→

Original post →

More from Fun

Fun channel →