避免由于对AI系统的不完全了解而产生的负面影响
Avoiding Negative Side Effects due to Incomplete Knowledge of AI Systems
摘要:
在现实世界中行动的自主智能体通常基于忽略环境的某些方面的模型进行操作。任何给定模型的不完整性-手工制作或机器获取-是不可避免的,因为任何建模技术对于复杂的现实世界设置的实际限制。由于其模型的有限保真度,代理的行为可能会在执行过程中产生意想不到的不良后果。学习识别和避免代理行为的这种负面影响对于提高自治系统的安全性和可靠性至关重要。减轻负面副作用是一个新兴的研究课题,由于人工智能系统的部署及其广泛的社会影响的快速增长,该课题正引起越来越多的关注。本文提供了一个全面的概述不同形式的负面影响和最近的研究努力,以解决这些问题。我们确定了负面副作用的关键特征,强调了避免负面副作用的挑战,并讨论了最近开发的方法,对比了它们的优点和局限性。文章最后讨论了开放的问题和建议,为未来的研究方向。
英文摘要:
Autonomous agents acting in the real-world often operate based on models that ignore certain aspects of the environment. The incompleteness of any given model -- handcrafted or machine acquired -- is inevitable due to practical limitations of any modeling technique for complex real-world settings. Due to the limited fidelity of its model, an agent's actions may have unexpected, undesirable consequences during execution. Learning to recognize and avoid such negative side effects of an agent's actions is critical to improve the safety and reliability of autonomous systems. Mitigating negative side effects is an emerging research topic that is attracting increased attention due to the rapid growth in the deployment of AI systems and their broad societal impacts. This article provides a comprehensive overview of different forms of negative side effects and the recent research efforts to address them. We identify key characteristics of negative side effects, highlight the challenges in avoiding negative side effects, and discuss recently developed approaches, contrasting their benefits and limitations. The article concludes with a discussion of open questions and suggestions for future research directions.
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。