按年被提到
20172017 · 2 次2026
数学、逻辑与理论›AI 数学与定理证明·arXiv:1706.08605 · cs.SE

Developing Bug-Free Machine Learning Systems With Formal Mathematics

Daniel Selsam, Percy Liang, David L. Dill · 2017
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2017-06-28 · 最近 2017-12-31
2不同的人
2活跃月份
2讨论串
39被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Noisy data, non-convex objectives, model misspecification, and numerical instability can all cause undesired behaviors in machine learning systems. As a result, detecting actual implementation errors can be extremely difficult. We demonstrate a methodology in which developers use an interactive proof assistant to both implement their system and to state a formal theorem defining what it means for their system to be correct. The process of proving this theorem interactively in the proof assistant exposes all implementation errors since any error in the program would cause the proof to fail. As a case study, we implement a new system, Certigrad, for optimizing over stochastic computation graphs, and we generate a formal (i.e. machine-checkable) proof that the gradients sampled by the system are unbiased estimates of the true mathematical gradients. We train a variational autoencoder using Certigrad and find the performance comparable to training the same model in TensorFlow.

候选层:至少 2 个不同的人、在 2 个不同月份提到过,还没到主书架的 3 人 3 个月。没有做立场标注,所以只写“被提及”。

HN 书评

推荐在前,同类按时间 · 每篇最多保留 20 条,统计用全部提及
评论cf · 2017-12

“I think this is a good application paper for Lean showing how you can use it to develop machine learning algorithms. https://arxiv.org/abs/1706.08605”

Coming Soon: Machine-Checked Proofs in Everyday Development原评论 ↗
被提交为 story · 1 次