Proceedings of the Detection and Classification of Acoustic Scenes and Events 2018 Workshop (DCASE2018)

The successful application of modern deep neural networks is heavily reliant on the chosen architecture and the selection of the appropriate hyperparameters. Due to the large number of parameters and the complex inner workings of a neural network, finding a suitable configuration for a respective problem turns out to be a rather complex task for a human. In this paper we, propose an evolutionary approach to automatically generate a suitable neural network architecture and hyperparameters for any given classification problem. A genetic algorithm is used to generate and evaluate a variety of deep convolutional networks. We take the DCASE 2018 Challenge as an opportunity to evaluate our algorithm on the task of acoustic scene classification. The best accuracy achieved by our approach was 74.7% on the development dataset.

[1]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[2]  Peter M. Todd,et al.  Designing Neural Networks using Genetic Algorithms , 1989, ICGA.

[3]  Thomas Bäck,et al.  The Interaction of Mutation Rate, Selection, and Self-Adaptation Within a Genetic Algorithm , 1992, PPSN.

[4]  Thomas Bäck,et al.  An Empirical Study on GAs "Without Parameters" , 2000, PPSN.

[5]  Risto Miikkulainen,et al.  Evolving Neural Networks through Augmenting Topologies , 2002, Evolutionary Computation.

[6]  Minyoung Kim,et al.  Deep Clustered Convolutional Kernels , 2015, FE@NIPS.

[7]  Wojciech Zaremba,et al.  An Empirical Exploration of Recurrent Network Architectures , 2015, ICML.

[8]  Colin Raffel,et al.  librosa: Audio and Music Signal Analysis in Python , 2015, SciPy.

[9]  Tuomas Virtanen,et al.  TUT database for acoustic scene classification and sound event detection , 2016, 2016 24th European Signal Processing Conference (EUSIPCO).

[10]  Kenneth O. Stanley,et al.  Simple Evolutionary Optimization Can Rival Stochastic Gradient Descent in Neural Networks , 2016, GECCO.

[11]  David Pfau,et al.  Convolution by Evolution: Differentiable Pattern Producing Networks , 2016, GECCO.

[12]  Ankit Shah,et al.  DCASE2017 Challenge Setup: Tasks, Datasets and Baseline System , 2017, DCASE.

[13]  Mark D. Plumbley,et al.  Neuroevolution for sound event detection in real life audio: A pilot study , 2017 .

[14]  Quoc V. Le,et al.  Large-Scale Evolution of Image Classifiers , 2017, ICML.

[15]  Valery Naranjo,et al.  EvoDeep: A new evolutionary approach for automatic Deep Neural Networks parametrisation , 2018, J. Parallel Distributed Comput..

[16]  Tuomas Virtanen,et al.  A multi-device dataset for urban acoustic scene classification , 2018, DCASE.