Bean Machine for AI Uncertainty by Meta

Dmitry Spodarets

Meta (meaning Facebook) made an announcement of Bean Machine. It is a probabilistic programming system that, formally, makes it easier to perform and learn about confusions in AI models. Bean Machine can be used to discover unobserved properties of a model via automatic, “uncertainty-aware” learning algorithms.

Probabilistic modeling, the kind of the AI technique that Bean Machine adopts, can identify these ambivalences by taking into account the impact of accidental events in predicting the instance of future outcomes. Likewise, this programming system is built on top of Meta’s PyTorch machine learning framework and Bean Machine Graph (BMG), a custom C++ backend.