Pieter Abbeel Weighs In on Humanoid Robot Training as Authors Show Real-Hardware Validation
pabbeel · x · 2026-09-06
UC Berkeley professor Pieter Abbeel asked @SongShuran whether training a single robot across enough tasks could generalize toward humanoids, calling one example "a pretty convincing stochastic gradient step" in that direction. The authors replied that, at reviewer 2's request, they also validated the design on real-world hardware.
Related event: Model auto-optimizes ALOHA robot design, validated on hardware(3 posts)→
More from Embodied
- Rivian acquires world model startup Third Dimension to accelerate its self-driving push — emmanuelvivier · 2026-10-23
- Nvidia's Berlin GTC spotlights humanoid robots and Vera Rubin AI supercomputing — emmanuelvivier · 2026-10-23
- XPeng commissions 80%-automated line for IRON humanoid, mass production by end-2026 — ingliguori · 2026-10-23
- HKU's TILDE Distills Embodied Foundation Models into Robot Policies — hkuhk · 2026-10-23
- 10 of 12 Beta Robots Worked Without Fine-Tuning — Do We Still Need It? — micoolcho · 2026-10-23
- EmbodiedEvo lets robot agents self-train, banking reusable Skills and Lessons across tasks — jiqizhixin · 2026-10-23