Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes — mentioned by jah242

Sat, 16 Apr 2022 18:26:26 GMT

This would be one example from Deepmind using raw pixel input to stack objects. This has a relatively detailed reward function (but is also a very complicated task) - https://arxiv.org/abs/2110.06192 There are other examples from OpenAI a while back using even just sparse rewards (i.e binary 1, 0 for success or failure over the whole task) - but these weren't pixel input if I remember correctly - https://openai.com/blog/ingredients-for-robotics-research/ I m afraid if you think providing any reward function is cheating then we have fundamentally different views of what AI/ML even…

Paper and discussion history. A mention is not an endorsement.

Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes · HackerPapers discussions

Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapes — mentioned by jah242