积极避免生成模型中的废话
Actively Avoiding Nonsense in Generative Models
摘要：
当生成模型适合最大化观察数据的可能性时，它可能会生成完全无意义的东西。这是由于“模型误差”，即，当真实数据生成分布不适合正在学习的生成模型的类时。为了解决这个问题，我们提出了一个主动分布学习的模型，使用二进制无效预言机，识别一些例子明显无效，以及从真实分布中采样的随机阳性例子。我们的目标是最大化的可能性，积极的例子受到约束的（几乎）永远不会产生的例子标记为无效的预言。与一类概率分布相比，担保是不可知的。我们表明，虽然适当的学习往往需要指数许多查询的无效预言，不适当的分布学习可以使用多项式许多查询。
英文摘要：
A generative model may generate utter nonsense when it is fit to maximize the likelihood of observed data. This happens due to "model error," i.e., when the true data generating distribution does not fit within the class of generative models being learned. To address this, we propose a model of active distribution learning using a binary invalidity oracle that identifies some examples as clearly invalid, together with random positive examples sampled from the true distribution. The goal is to maximize the likelihood of the positive examples subject to the constraint of (almost) never generating examples labeled invalid by the oracle. Guarantees are agnostic compared to a class of probability distributions. We show that, while proper learning often requires exponentially many queries to the invalidity oracle, improper distribution learning can be done using polynomially many queries.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
