Connectionists: A summer school on the intersection between neuroscience and AI

[Danko Nikolic]

I am organizing a summer school entitled “Intersection Between Neuroscience
and AI”. The summer school will take place on the beautiful Croatian coast,
likely in September 2025.


About the summer school:


The contents will be heavily biased towards three closely related topics i)
practopoiesis, ii) transient selection of subnetworks, and iii) gating in
machine learning. In addition, a lot of other topics in neuroscience and AI
will be covered.


Practopoiesis is a theory of hierarchical adaptation:
https://shorturl.at/pTIUE


Transient selection of subnetworks is a hypothesis that the metabotropic
receptors and G protein-gated ion channels drive our cognition:
https://shorturl.at/gubr3


Gating proposes that we need to add gates to artificial neural networks for
AI to achieve human-level intelligence: https://shorturl.at/GRvee


If you want to learn more about those topics, this summer school is the
right opportunity.


Here is the tentative curriculum:


Introduction to common knowledge in neuroscience and cognitive science:
- Neurons, synapses, dendrites, axons, action potentials, short-term
plasticity, long-term plasticity

- Working memory, attention, grouping and binding in perception, concepts
and schemas

- The 4E problem of robotics

- The basics of deep learning

- The big three: the three fundamental problems with current AI: Scaling,
Accumulation of knowledge, Novel situations. - The problem of consciousness


Fundamental adaptive theory:
- The concept of adaptive hierarchy

- The concept of achieving intelligence through poiesis

- The concept of inductive bias

- The concept of an adaptive traverse of knowledge

- Free energy principle


Biological adaptive theory:
- Traverses of a single cell

- Traverses in the existing brain theories

- Why thinking with two traverses is better than with one?

- Theory of evolution and adaptive traverses

- The adaptive role of metabotropic receptors and G protein-gated ion
channels


Implications for artificial intelligence:
- Symbolic AI and traverses of knowledge?

- Why does deep learning have an adaptive hierarchy of only two traverses?

- Gating as a new paradigm for AI

- Why gating requires shallow rather than deep neural networks

- Gating enables inference with two traverses

- No free lunch in machine learning

- Learning by accumulating knowledge


Several speakers have already confirmed participation.


The summer school is planned to take place in the Hotel Soline in the
beautiful Brela, Croatia. The discounted price for single occupancy is €215
and for double occupancy is €240, which includes breakfast.


In addition, we are considering €500 for the registration fee. We may have
fee waivers for students. We are working on getting a sponsorship.


At this stage, we need to assess the interest. To express interest, reserve
a spot, or contribute in some other way please write an email to:
danko at robotsgomental.com




Dr. Danko Nikolić
CEO, Robots Go Mental
www.robotsgomental.com
www.danko-nikolic.com
https://www.linkedin.com/in/danko-nikolic/
-- I wonder, how is the brain able to generate insight? --
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20240814/d7156481/attachment.html>