• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI
Announcement

AlphaNIST reportedly achieves 98.2% success across 20 NIST assembly tasks in simulation

A team member says one policy learned using a binary success-or-failure reward and a simple control cost.

Octi ZhangOZ
1 Source, 9h ago, first seen 9h ago

TLDR

An AlphaNIST team member says a single policy, trained from scratch with a binary success-or-failure reward and a simple control cost, achieved a 98.2% success rate across all 20 NIST assembly tasks in simulation. They say it can also complete 10 different tasks in a row, receiving the task reward only after all 10 are done.

Combined views

92.6K

1 Source, first seen 9h ago

938 likes22 comments715 saves118 reposts

Combined views

92.6K

1 Source, first seen 9h ago

938 likes22 comments715 saves118 reposts
Image: @octi_zhang on X

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Featured Source

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

1 Source

Octi Zhang@octi_zhang(1/6) From @GoogleDeepmind’s AlphaGo to AlphaZero, we saw machines find moves that reached beyond what we’d imagined for them. No one taught them those strategies. They emerged from reinforcement learning: playing, failing, and trying again under a single binary task reward for success or failure. Inspired by them, we couldn’t help but wonder: what might a robot discover if we gave it that same freedom to learn? That question led us to AlphaNIST. Starting tabula rasa, with that same binary reward plus only a simple control cost, a single policy solves all 20 @NIST assembly tasks at a 98.2% success rate in simulation. More surprising still, AlphaNIST can complete 10 different tasks in a row, receiving that reward only once, after all 10 are done. Just as AlphaGo’s strategies surprised its creators, AlphaNIST discovered dexterous strategies we couldn’t have come up with ourselves. Paper: https://arxiv.org/abs/2610.13354 Website: http://alpha-nist.github.io9h
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    1 Source

    Octi Zhang@octi_zhang(1/6) From @GoogleDeepmind’s AlphaGo to AlphaZero, we saw machines find moves that reached beyond what we’d imagined for them. No one taught them those strategies. They emerged from reinforcement learning: playing, failing, and trying again under a single binary task reward for success or failure. Inspired by them, we couldn’t help but wonder: what might a robot discover if we gave it that same freedom to learn? That question led us to AlphaNIST. Starting tabula rasa, with that same binary reward plus only a simple control cost, a single policy solves all 20 @NIST assembly tasks at a 98.2% success rate in simulation. More surprising still, AlphaNIST can complete 10 different tasks in a row, receiving that reward only once, after all 10 are done. Just as AlphaGo’s strategies surprised its creators, AlphaNIST discovered dexterous strategies we couldn’t have come up with ourselves. Paper: https://arxiv.org/abs/2610.13354 Website: http://alpha-nist.github.io9h
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet