We, as a community composed of neuroscientists, neuromorphic engineers, roboticists, and theoreticians have been trying to understand neural processing and computation from different perspectives. Our aim is to figure out the key elements which bootstrap the brain to be the unique computing machine it is, and to enable neuromorphic intelligence.
These elements (genes, proteins, neurons, or synapses), aggregated in sub-modules (gene regulatory networks, nuclei, layered ensembles) and their interfaces (message-parsing systems, connectivity schemes, mechanism of plasticity) might hint their functional role in neural computation.
It is, however, their modularity and composability which provides a framework for constructing neuromorphic processing systems.
The modularity forces us to think of modular computation and construct modules, which consist of sub-modules, thus inherently requiring a sense of hierarchy. The composability, on the other hand, highlights the need for the re-usability of said modules to establish a decentralized orchestration of computation across scales of complexity and time.
Last year, we focused on how to connect various building blocks, the pieces of the puzzle. This year’s Capocaccia workshop aims to use this knowledge to solve real-world problems that animals encounter in their daily lives such as defining goals, performing inference in a noisy environment or representing and storing sequences. Our goal is to identify the operational definitions of computation, behavior, and intelligence relevant to solving everyday problems.
The workshop features open and highly interactive discussion sessions in the morning; hands-on projects, tutorials, and hardware and software jamming sessions during the day; and free-form discussions in the evenings.
The workshop is open to everyone, but since resources are limited, we can accept only a limited number of registrations. Due to the limited number of hotel rooms, Ph.D. students are expected to pair up and share rooms. Participants and invited discussion leaders are encouraged to stay for the full two week period, but can stay for less if necessary. They are however expected to use their own funding to pay for the registration fee, the travel and accommodation expenses.
Chiara Bartolozzi (Italian Institute of Technology, Genova, Italy)
Florian Engert (Department of Molecular and Cellular Biology, Harvard University, Boston, USA)
Georg Keller (Friederich Miescher Institute, Basel, Switzerland)
Germain Haessig (Division of Computational Science and Technology, Karolinska Institute, Stockholm, Sweden)
Melika Payvand (Institute of Neuroinformatics, ETHZ & UZH, Zurich, Switzerland)
Moritz Milde (International Centre for Neuromorphic Systems, Western Sydney University, Sydney, Australia)
Yves Fregnac (Paris Saclay Insitute of Neuroscience, France, Paris)