μneuron | Synthetist Meetups
DATE: Thursday, 30 May 2024 | 17:00-20:00
LOCATION: osmo/za, Ljubljana
MENTORS: Staš Vrenko & Gregor Krpič
PARTICIPATION: The workshop is fully booked. The participation fee, which covers part of the material costs of the device, is 20€ and should be paid at the workshop. The device is yours to keep.
The meetup will explore the functioning of artificial neurons and neural networks through the lens of analogue computing, which has gained renewed relevance in the context of the analogue revolution in machine learning.
We will assemble the μneuron, which serves as an analogue neural network model and an interesting tool for processing audio, logical (gates, triggers, etc.) and control (CV) signals. The μneuron instrument consists of three analogue neurons and an interface allowing users to modify their connections (topologies) and manually tune their weights.
The mentors developed the instrument based on one of the pioneering electronic circuits in biological neuron modelling, the so-called Harmon Preliminary Neuron Model, which dates back to the 1960s. While the computing power of the three neurons cannot match the intricate workings of digital neural networks, such physical models can provide an in-depth understanding of how neural nets operate.
Gallery
Mentorja
Staš Vrenko is an artist, musician and designer of electronic instruments. His artistic practice combines different fields of art with a focus on exploring sound, electronic media and fabrication technologies. To date, he has presented several solo projects and has exhibited in many international festivals and group exhibitions.
Gregor Krpič explores sound through the development of his sensory electronic instruments. He has been actively involved in various contemporary investigative art projects for many years. He has been collaborating in developing and mentoring workshops within the Rampa project at the Kersnikova Institute since 2018.
Credits
Supported by the Municipality of Ljubljana and the Public Fund for Cultural Activities (JSKD).