T. Nakamura

Tadahiro Taniguchi, Shingo Murata, Masahiro Suzuki et al.

Creating autonomous robots that can actively explore the environment, acquire knowledge and learn skills continuously is the ultimate achievement envisioned in cognitive and developmental robotics. Their learning processes should be based on interactions with their physical and social world in the manner of human learning and cognitive development. Based on this context, in this paper, we focus on the two concepts of world models and predictive coding. Recently, world models have attracted renewed attention as a topic of considerable interest in artificial intelligence. Cognitive systems learn world models to better predict future sensory observations and optimize their policies, i.e., controllers. Alternatively, in neuroscience, predictive coding proposes that the brain continuously predicts its inputs and adapts to model its own dynamics and control behavior in its environment. Both ideas may be considered as underpinning the cognitive development of robots and humans capable of continual or lifelong learning. Although many studies have been conducted on predictive coding in cognitive robotics and neurorobotics, the relationship between world model-based approaches in AI and predictive coding in robotics has rarely been discussed. Therefore, in this paper, we clarify the definitions, relationships, and status of current research on these topics, as well as missing pieces of world models and predictive coding in conjunction with crucially related concepts such as the free-energy principle and active inference in the context of cognitive and developmental robotics. Furthermore, we outline the frontiers and challenges involved in world models and predictive coding toward the further integration of AI and robotics, as well as the creation of robots with real cognitive and developmental capabilities in the future.

Tadahiro Taniguchi, Hiroshi Yamakawa, Takayuki Nagai et al.

Building a humanlike integrative artificial cognitive system, that is, an artificial general intelligence (AGI), is the holy grail of the artificial intelligence (AI) field. Furthermore, a computational model that enables an artificial system to achieve cognitive development will be an excellent reference for brain and cognitive science. This paper describes an approach to develop a cognitive architecture by integrating elemental cognitive modules to enable the training of the modules as a whole. This approach is based on two ideas: (1) brain-inspired AI, learning human brain architecture to build human-level intelligence, and (2) a probabilistic generative model(PGM)-based cognitive system to develop a cognitive system for developmental robots by integrating PGMs. The development framework is called a whole brain PGM (WB-PGM), which differs fundamentally from existing cognitive architectures in that it can learn continuously through a system based on sensory-motor information. In this study, we describe the rationale of WB-PGM, the current status of PGM-based elemental cognitive modules, their relationship with the human brain, the approach to the integration of the cognitive modules, and future challenges. Our findings can serve as a reference for brain studies. As PGMs describe explicit informational relationships between variables, this description provides interpretable guidance from computational sciences to brain science. By providing such information, researchers in neuroscience can provide feedback to researchers in AI and robotics on what the current models lack with reference to the brain. Further, it can facilitate collaboration among researchers in neuro-cognitive sciences as well as AI and robotics.

Wataru Takase, Tomoaki Nakamura, Yoshiyuki Watase et al.

Monitoring is indispensable to check status, activities, or resource usage of IT services. A combination of Kibana and Elasticsearch is used for monitoring in many places such as KEK, CC-IN2P3, CERN, and also non-HEP communities. Kibana provides a web interface for rich visualization, and Elasticsearch is a scalable distributed search engine. However, these tools do not support authentication and authorization features by default. In the case of single Kibana and Elasticsearch services shared among many users, any user who can access Kibana can retrieve other's information from Elasticsearch. In multi-user environment, in order to protect own data from others or share part of data among a group, fine-grained access control is necessary. The CERN cloud service group had provided cloud utilization dashboard to each user by Elasticsearch and Kibana. They had deployed a homemade Elasticsearch plugin to restrict data access based on a user authenticated by the CERN Single Sign On system. It enabled each user to have a separated Kibana dashboard for cloud usage, and the user could not access to other's one. Based on the solution, we propose an alternative one which enables user/group based Elasticsearch access control and Kibana objects separation. It is more flexible and can be applied to not only the cloud service but also the other various situations. We confirmed our solution works fine in CC-IN2P3. Moreover, a pre-production platform for CC-IN2P3 has been under construction. We will describe our solution for the secure use of Kibana and Elasticsearch including integration of Kerberos authentication, development of a Kibana plugin which allows Kibana objects to be separated based on user/group, and contribution to Search Guard which is an Elasticsearch plugin enabling user/group based access control. We will also describe the effect on performance from using Search Guard.