New year’s greetings from the Whole Brain Architecture Initiative (WBAI)!
Last year, the Whole Brain Architecture Initiative (WBAI) continued its activities in education and R&D with the support and cooperation of many, including supporting members, the WBA Seminar Committee (formerly the WBA Supporters), SIG-WBA, advisors, and collaborating organizations, while in the midst of the COVID-19 pandemic.
As one of the major activities, we held the 5th WBA Symposium under the theme of “the revealed target of intelligence research.” In Part I, the WBA approach that has evolved over the past five years was explained. In Part II, the WBAI Incentive Awards were presented to Matthew Crosby and Benjamin Beyret, who organized the Animal AI Olympics to promote R&D to reproduce common sense reasoning in animals, and Masahiro Suzuki, who contributed to the R&D of deep generative models and thus to the foundation of brain-inspired artificial intelligence. The Award for Meritorious Service for the NPO was also presented to Mr. Takahiro Ikushima. In Part III, the main target of intelligence research, approaches for it, and how brain knowledge can be utilized in the process were discussed.
WBA seminars were also held with the themes of “Brain and Creativity,” “AGI and Symbiotic Interaction,” and “the Predicting Brain and Phenomenology of Subjectivity.” Prior to the 5th WBA Hackathon to be held this year, we held a modelathon to solicit models of working memory.
WBAI’s activities to support R&D have been divided into activities to create Brain Reference Architecture (BRA) that accumulates neuroscience knowledge in a form useful for the development of brain-inspired AGI, and technologies to utilize BRA to promote the implementation of brain-inspired software.
Our BRA creation activities in 2020 include modeling the ability of neocortical local circuits to select behavioral outputs generated by the thalamic relay cell de-inhibition in the basal ganglia [1], and locating the function of pathway integration, which is essential for navigation, in the second layer of the olfactory cortex [2]. We also made computational meanings of signals exchanged between neocortical areas [3], and the role of the Papez circuit in introspection [4]. Furthermore, we created a database to register BRA and visualize its contents for the development of brain-inspired AGI [5].
Through these experiences, WBAI is accumulating the know-how to create BRAs, though we need to expand the activity further to construct BRA for the entire brain.
In 2021, we will continue our efforts in research promotion and human resource development projects in the direction of expanding our activities to create BRAs. We thank you for your continued support and patronage!
January 2021
Members of the Whole Brain Architecture Initiative
References:
- [1] Yamakawa, H. (2020). Attentional Reinforcement Learning in the Brain. New Generation Computing. https://doi.org/10.1007/s00354-019-00081-z
- [2] Fukawa, A., Aizawa, T., Yamakawa, H., & Yairi, I. E. (2020). Identifying Core Regions for Path Integration on Medial Entorhinal Cortex of Hippocampal Formation. Brain Sciences, 10(1). https://doi.org/10.3390/brainsci10010028
- [3] Yamakawa, H. (2020). Revealing the computational meaning of neocortical interarea signals. Frontiers in Computational Neuroscience, 14, 74.
- [4] Yamakawa, H. (2020). The Role of the Papez Circuit for Introspection, Special Interest Group on Artificial General Intelligence, Japanese Society of Artificial Intelligence, SIG-AGI-015-03. (in Japanese) http://id.nii.ac.jp/1004/00010725/
- [5] Sasaki, M., Arakawa, N., & Yamakawa, H. (2020). Constructing Structured Data on Functional Hypotheses of the Brain. Poster at International Symposium on Artificial Intelligence and Brain Science. http://www.brain-ai.jp/wp-content/uploads/2020/10/PDF_31.pdf
Japanese