تقرير
Experiments with Detecting and Mitigating AI Deception
| العنوان: | Experiments with Detecting and Mitigating AI Deception |
|---|---|
| المؤلفون: | Sahbane, Ismail, Ward, Francis Rhys, Åslund, C Henrik |
| سنة النشر: | 2023 |
| المجموعة: | Computer Science |
| مصطلحات موضوعية: | Computer Science - Artificial Intelligence |
| الوصف: | How to detect and mitigate deceptive AI systems is an open problem for the field of safe and trustworthy AI. We analyse two algorithms for mitigating deception: The first is based on the path-specific objectives framework where paths in the game that incentivise deception are removed. The second is based on shielding, i.e., monitoring for unsafe policies and replacing them with a safe reference policy. We construct two simple games and evaluate our algorithms empirically. We find that both methods ensure that our agent is not deceptive, however, shielding tends to achieve higher reward. Comment: 4 pages, 2 figures, 3 algorithms, 1 table |
| نوع الوثيقة: | Working Paper |
| URL الوصول: | http://arxiv.org/abs/2306.14816 |
| رقم الأكسشن: | edsarx.2306.14816 |
| قاعدة البيانات: | arXiv |
| الوصف غير متاح. |