按年被提到
20212021 · 3 次2026
机器学习与深度学习›机器学习实践与反思·arXiv:2109.13916 · cs.LG

Unsolved Problems in ML Safety

Dan Hendrycks, Nicholas Carlini, John Schulman, Jacob Steinhardt · 2021
开始阅读 ↗PDFalphaXiv 讨论
被提及 · 2 人首次 2021-09-29 · 最近 2021-10-09
2不同的人
2活跃月份
3讨论串
217被引(OpenAlex)
arXiv 摘要(候选层不写导读)

Machine learning (ML) systems are rapidly increasing in size, are acquiring new capabilities, and are increasingly deployed in high-stakes settings. As with other powerful technologies, safety for ML should be a leading research priority. In response to emerging safety challenges in ML, such as those introduced by recent large-scale models, we provide a new roadmap for ML Safety and refine the technical problems that the field needs to address. We present four problems ready for research, namely withstanding hazards ("Robustness"), identifying hazards ("Monitoring"), reducing inherent model hazards ("Alignment"), and reducing systemic hazards ("Systemic Safety"). Throughout, we clarify each problem's motivation and provide concrete research directions.

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

没有评论直接链到它;下面是它被提交的讨论串里回复最多的评论(不计入统计)
讨论串评论unholiness · 2021-10

“Some overlap here with the 2017 paper from Google Brain et al: Concrete Problems in AI Safety: https://arxiv.org/pdf/1606.06565.pdf A nice video from the wonderful AI safety communicator Robert Miles outlines those problems: https://www.youtube.com/watch?v=AjyM-f8rDpg”

Unsolved Problems in ML Safety原评论 ↗
讨论串评论sgt101 · 2021-10

“These are unsolved in the sense that there is no formal or comprehensive solution to them, and in the limit that means that ML cannot be used for some applications. On the other hand there are mitigations that can be implemented to increase the confidence that we have in ML centric solutions - and that can bring some applications back into scope. Figure…”

Unsolved Problems in ML Safety原评论 ↗
被提交为 story · 3 次