AlphaNIST：仅用二值奖励让机器人学会 NIST 装配板 20 项任务
AlphaNIST: Emergent Assembly Skills from Binary Task Rewards
学 AlphaGo Zero 的思路做机器人装配：不给演示、只给成败信号，一个策略搞定 NIST 装配板全部 20 个任务，成功率达 98.2%，值得看看它自己摸索出了什么招。
arXiv 论文 AlphaNIST 探索机器人能否只靠稀疏的二值任务奖励学会装配，不使用演示、进度塑形奖励或脚本化原语。在仿真中，单一策略从零开始训练，覆盖 NIST Assembly Task Board #1 的全部 20 项装配任务，总体训练成功率达 98.2%。该框架还能在单个回合内完成 10 对装配，且不规定顺序，只在全部 10 对同时完成时才给奖励。学到的策略表现出灵巧和规划式行为，例如为后续装配选择抓取方式、利用周围几何结构翻转零件、先准备一个零件再去装另一个再回来收尾，全程没有显式规划器。
AlphaNIST: Emergent Assembly Skills from Binary Task Rewards
The NIST Assembly Task Board #1 remains a challenging benchmark for learning-based methods in robotic manipulation. Over short horizons, it demands precise, contact-rich control of small, tight-fitting parts. Over long horizons, it requires recovery and coordination across extended action sequences. Demonstrations, task decomposition, and dense progress rewards have enabled research advances by providing guidance on how to assemble. But how much of the strategy must be specified? The progression from AlphaGo to AlphaGo Zero showed that RL at scale can discover superhuman strategies without prescribing how to win. AlphaNIST asks whether a sparse binary task reward and task diversity admitted by this task-agnostic reward can similarly let a robot discover how to assemble. In simulation, we train a single policy from scratch across all 20 assembly tasks on the NIST Assembly Task Board #1, reaching 98.2% overall training success without demonstrations, progress shaping, scripted primitives, or per-task tuning. The framework also trains policies to complete ten assembly pairs in a single episode without a prescribed order, with task success rewarded only when all ten pairs are simultaneously complete. The learned policies discover dexterous and planning-like behaviors: choosing grasps for later assembly, using surrounding geometry to reorient parts, and preparing one part, assembling another, then returning to finish the first without an explicit planner. For videos, see our website: https://alpha-nist.github.io/
