Losses, Dissonances, and Distortions

TL;DR

This paper explores a novel method where training losses and gradients influence musical dissonance and visual distortions in a live performance, creating a feedback loop between machine learning training and artistic expression.

Contribution

It introduces a system that integrates training dynamics into live musical and visual performances, enabling performer-controlled feedback loops.

Findings

Losses and gradients generate artistic dissonance and distortion effects.

The system allows performer influence over the training process.

Creates a closed feedback loop between machine learning and live art.

Abstract

In this paper I present a study in using the losses and gradients obtained during the training of a simple function approximator as a mechanism for creating musical dissonance and visual distortion in a solo piano performance setting. These dissonances and distortions become part of an artistic performance not just by affecting the visualizations, but also by affecting the artistic musical performance. The system is designed such that the performer can in turn affect the training process itself, thereby creating a closed feedback loop between two processes: the training of a machine learning model and the performance of an improvised piano piece.