Synapses and Silicon: Exploring the Intersection of AI and Neuroscience

EAISI AI Symposium

Tuesday March 28, 2023 from 12:30 PM to 2:30 PM
Neuron building, TU/e campus

On Tuesday March 28th 2023, EAISI organizes an AI symposium in honor of the official opening of the new building Neuron which will house the Eindhoven AI Systems Institute but also is a large education and study facility at the TU/e campus. Lunch is provided for participants from 11.30 - 12.30h at the EAISI lounge in Neuron 1.116.

The official opening will take place from 15.00-16.00h, in presence of Constantijn van Oranje, special envoy

Dominique Fürst and Carlo van de Weijer




11.30 - 12.30        Lunch for participants AI symposium | EAISI lounge, first floor Neuron 1.116

--- The AI symposium takes place at the atrium in Neuron ---




12:30 - 12.40              

Carlo van de Weijer

General Manager of EAISI 



12.40 - 13.05

Regina Luttge and Bert de Vries
Associate Professor Microsystems | Full Professor Signal Processing Systems, TU/e

BayesBrain: The World’s First Brain-on-Chip AI Computer

13.05 - 13.30 Véronne Reinders
CTO Aristotle Technologies
Train your brain with Aristotle Technologies
13.30 - 13.55 Bouke van Balen
PhD candidate Philosophy & Ethics of Brain-Computer Interfaces, TU/e, UMC U, TU Delft
Ethical issues of using A.I. in Brain-Computer Interfaces
13.55 - 14.20 Yoeri van de Burgt and Imke Krauhausen
Associate Professor Microsystems  | PhD candidate at Neuromorphic Engineering group, TU/e

Plastic electronics for artificial synapses and neurons

14.20 - 14.30 Closing

Chair: Dominique Fürst

BayesBrain: The World’s First Brain-on-Chip AI Computer

Regina Luttge and Bert de Vries

Computation in biological brain tissue consumes several orders of magnitude less power than silicon-based systems.

Motivated by this fact, this project aims to develop the world’s first hybrid neuro-in-silico Artificial Intelligence (AI) computer, introducing a fundamentally new paradigm of AI computing.

In this high-risk high-gain project, we will combine an in-silico Bayesian control agent (BCA) with neural tissue hosted by a microfluidic Brain-on-Chip (BoC) that together form a hybrid learning system capable of solving real-world AI problems.

Toward this paradigm, all computation and communication inside and between the BCA and BoC will be governed by the Free Energy Principle, which is both the leading neuroscientific theory for describing biological neuronal processes and supports a variational Bayesian machine learning interpretation.

We will start by developing a pure silicon-based BCA that learns to balance an inverted pendulum, implemented by free energy minimization on a factor graph.

Next, we will replace successively larger parts of the factor graph with biological neural circuits of a microfluidic multi-compartment BoC device. The biological network will be trained by electrical stimulation orchestrated by the synthetic Bayesian agent.

For the communication between these two units, we will design and realize a novel communication protocol making use of existing software being applied in readout and event sorting for Calcium imaging and multi-electrode array data, such as MEAViewer, CALIMA, NetCal and SpikeHunter.

By upscaling the number of replaced sub-circuits, we aim to provide a proof-of-concept and to lay the basis for ultra-low power hybrid brain- on-chip AI computing.

Train your brain with Aristotle Technologies

Véronne Reinders

Every 9 seconds someone’s life changes completely because of an acquired brain injury.

At Aristotle Technologies we develop evidence-based and data-driven software to train your brain.
It is our aim to support, challenge, and engage individuals in their cognitive development and make life a bit more bearable for those with an acquired brain injury, but also for congenital brain injuries and for those who simply benefit from training their cognition.

We all try to stay physically fit in our own way, so why not do the same for the most complex and important part of our body and train our brain?

Ethical issues of using A.I. in Brain-Computer Interfaces

Bouke van Balen

Neurotechnologies and Artificial Intelligence (AI) are becoming more intertwined, and this raises unique ethical issues.

In this talk, I will highlight some ethical issues that surface when Brain Computer Interfaces (BCIs) are used as assistive communication technologies for people with Locked-In Syndrome (LIS). Brain Computer Interfaces are devices that can be controlled with brain activity. This can be an outcome for people with LIS, who are fully paralyzed yet cognitively intact, and suffer from the inability to express themselves.

With BCI-technology, people with LIS can control devices that allow them to communicate without the usual need for muscle control. Promising research is being done into a specific type of implanted BCI that directly translates neural activity associated with attempted speech into words on a screen or word pronunciation by a synthesized voice.

With this BCI, people with LIS could get a method of communication that approaches conventional speech. AI is an important part of this technology, as it is used to decode brain signals and predict attempted speech.

Whereas such use of AI leads to efficient and faster speech, it raises questions about the agency and ownership of BCI-utterances. It can for instance be ambiguous if users have sufficient control over their BCI-utterances, and if these utterances represent them.

As it is important to have control over if we say something, what we say, and how we say it, these ethical issues deserve attention.

Plastic electronics for artificial synapses and neurons

Imke Krauhausen and Yoeri van de Burgt

Neuromorphic engineering takes inspiration from the efficiency of the brain and focusses on how to utilize its functionality such as neurons and synapses, in hardware. However, efficiently embedding artificial neural networks in hardware remains a significant challenge.

Due to their low power operation, easy tuneability and biocompatibility, organic electronic materials have shown potential to overcome some of these challenges, while simultaneously offer promising solutions for the manipulation and processing of biological signals and potential applications ranging from brain-computer-interfaces to adaptive sensing.

In this talk we demonstrate two device concepts based on novel organic mixed-conducting materials and show how we can use these devices as artificial neurons and synapses in smart autonomous robotics, trainable biosensors and sensory coding.