11 March 2017 | 12:00

Workshop Randomness & Machine Learning

This workshop given by Robert Lisek uses randomness and machine learning methods for creation of sound, visuals, performance, installation, interactive media, physical computing, and networking. This is practical course with the use of Supercollider, Pure data/Max Msp, Fluxus, Python, Lisp and hacking analogue devices for detection, creation and amplification of noise.

To obtain interesting results in music and art we need randomness. Randomness is important when you want the Neural Network from the same input to create different possibilities as an output, without generating the same output over and over again. Therefore it is different than in science, which is all about grouping and clustering. In art and music we don't want to endlessly obtain the same result. When you are composing sound or images, you don't want the neural network program to create the same sets of sounds and images; instead, you need creativity and variability. One of the solutions is to parametrize NN outputs with the use of probability distribution. A different way is to add noise directly to NN, instead of modeling the distribution parameters. Paradoxically, in such NNs, the more randomness during training, the better the results. Good random generators allow to avoid situations, when NN gets fixed in local minima.

The workshop consists of:

  • work with three main types of learning algorithms: Neural Networks and Deep Learning, Reinforcement Learning, Genetic Algorithms. Supervised and unsupervised learning, data sampling, backpropagation, explorations of environment, generalization, experimentation, markov models and processes (speech recognition), support vector machines (image recognition)

  • practical application of different types of random generators for constructing visual works, music compositions (random generators, random walks, monte carlo, etc), dealing with large systems of events: probability distributions (average, spread, deviation)

  • extraction of randomness from a physical processes through hacking analogue devices for detection, amplification, sonic and visual analysis (electromagnetic fields, gas particles, photons and decay of radioactive materials) 

  • applications RG and NN in building of new prototypes, artworks and network app Instructor

Robert B. LISEK
is an artist and mathematician who focuses on systems and processes (computational, biological, social). He is involved in the number of projects focused on radical art strategies, hacktivism and tactical media. Drawing upon conceptual art, software art and meta-media, his work intentionally defies categorization. Lisek is a pioneer of art based on AI and bioinformatics. Lisek is also a composer of contemporary music, author of many projects and scores on the intersection of spectral, stochastic, concret music, musica futurista and noise. Lisek is also a scientist who conducts a research in the area of theory of partially ordered sets in relation with artificial intelligence and machine learning.

He is the author of many exhibitions and concerts, among others: QUANTUM ENIGMA - Harvestworks Arts Center New York; RADICAL MIND - Columbia University New York, TERROR ENGINES - WORM Center Rotterdam; DEMONS - Venice Biennale (accompanying events); Manifesto vs. Manifesto - Ujazdowski Castel of Contemporary Art, Warsaw; NGRU - FILE, Sao Paulo; NEST - ARCO Art Fair, Madrid; Float - DMAC Harvestworks and Lower Manhattan Cultural Council, NYC; WWAI - Siggraph, Los Angeles; Falsecodes - Red Gate Gallery, Beijing; Gengine - National Gallery, Warsaw; Flextex - Byzantine Museum, Athens, FXT- ACA Media Festival, Tokyo and ISEA, Nagoya.
More information: www.fundamental.art.pl