Richard Sutton praises PhD thesis on robots that keep learning after deployment
RichardSSutton · x · 2026-10-04
Richard Sutton highlights Gautham Vasan's new PhD thesis "Robots That Learn on the Fly Through Real-World Interaction," which he says is exceptionally well done. Chapter 6 introduces AVG, a new streaming RL actor-critic algorithm distinct from the Stream-X family.
The thesis attacks the prevailing train-then-freeze paradigm in robot learning: policies are fixed after deployment, so degradations require engineers to collect data and retrain. It instead enables robots to learn from their own runtime experience, improving over their operational lifetime and adapting to unforeseen conditions.
More from Embodied
- 45 years of Marc Raibert: from one-legged hoppers to SoftBank buying his new lab twice — lukas_m_ziegler · 2026-11-20
- Meta to price Muse Charm AI device at $300 with up to a year of free Verizon 5G — Scobleizer · 2026-11-20
- Meta's Muse Charm AI gadget reportedly priced at $300, shipping by December — The Verge AI · 2026-11-20
- Nest doorbell catches a clawed robot autonomously tidying a yard — Buderus69 · 2026-11-20
- NVIDIA showcases developers building simulation apps with frontier AI agents like GPT-6 Astra — liu_mingyu · 2026-11-20
- NVIDIA and Microsoft bring Vera Rubin NVL72 to Azure and Nemotron 3.5 to Foundry — NVIDIA Blog · 2026-11-20