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
- Open-source ESP32 desk buddy shows what your Claude Code and Copilot agents are doing — DanWahlin · 2026-10-04
- Apple's home camera reportedly records no video, relies fully on AI sensing — ngxson · 2026-10-04
- Real estate mogul sells Bentley for Model Y, buying FSD Teslas for 10 employees — Vjeux · 2026-10-04
- Meta's Ray-Ban Gen 3 gets FDA hearing aid certification, with unexamined legal implications — thursdai_pod · 2026-10-04
- Musk envisions a robot buddy for everyone and 'universal high income' in 90% good outcome — heyshrutimishra · 2026-10-04
- Roboticist publishes full video analysis of Figure's F.02 decommissioning — MarwaEldiwiny · 2026-10-04