Recently deep learning based access control (DLBAC) model has been developed to reduce the burden of accesscontrol model engineering on a human administrator, while managing accurate accesscontrol state in large, co...
详细信息
ISBN:
(纸本)9798400705557
Recently deep learning based access control (DLBAC) model has been developed to reduce the burden of accesscontrol model engineering on a human administrator, while managing accurate accesscontrol state in large, complex, and dynamic systems. DLBAC utilizes neural networks for addressing accesscontrol requirements of a system based on user and resource metadata. However, in today's rapidly evolving, dynamic, and complex world with billions of connected users and devices, there are various environmental aspects in different application domains that affect accesscontrol rights and decisions. While Attribute-basedaccesscontrol (ABAC) have captured environmental factors through environmental attributes, DLBAC still lacks the capabilities of capturing any environmental factors and its use in accesscontrol decision making. In this paper, we propose an environment aware deep learning based access control model (DLBAC-Env) which includes environmental metadata in addition to user and resource metadata. We present an Industrial Internet of Things (IIoT) use case to demonstrate the need for DLBAC-Env and show how different types of environmental aspects in a specific domain are necessary towards making dynamic and autonomous accesscontrol decisions. We enhance the DLBAC model and dataset to incorporate environmental metadata and then implement and evaluate our DLBAC-Env model. We also present a reference implementation of DLBAC-Env in an edge cloudlet using AWS Greengrass.
Humanoid robots will be able to assist humans in their daily life, in particular due to their versatile action capabilities. However, while these robots need a certain degree of autonomy to learn and explore, they als...
详细信息
ISBN:
(纸本)9798400704918
Humanoid robots will be able to assist humans in their daily life, in particular due to their versatile action capabilities. However, while these robots need a certain degree of autonomy to learn and explore, they also should respect various constraints, for accesscontrol and beyond. We explore the novel field of incorporating privacy, security, and accesscontrol constraints with robot task planning approaches. We report preliminary results on the classical symbolic approach, deep-learned neural networks, and modern ideas using large language models as knowledge base. From analyzing their trade-offs, we conclude that a hybrid approach is necessary, and thereby present a new use case for the emerging field of neuro-symbolic artificial intelligence.
Humanoid robots will be able to assist humans in their daily life, in particular due to their versatile action capabilities. However, while these robots need a certain degree of autonomy to learn and explore, they als...
详细信息
ISBN:
(纸本)9798400701733
Humanoid robots will be able to assist humans in their daily life, in particular due to their versatile action capabilities. However, while these robots need a certain degree of autonomy to learn and explore, they also should respect various constraints, for accesscontrol and beyond. We explore incorporating privacy and security constraints (Activity-Centric accesscontrol and deep learning based access control) with robot task planning approaches (classical symbolic planning and end-to-end learning-based planning). We report preliminary results on their respective trade-offs and conclude that a hybrid approach will most likely be the method of choice.
暂无评论