Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Echo Networks, a novel recurrent neural network architecture, are introduced for event detection and classification tasks, particularly targeting resource-constrained edge applications. Unlike traditional feed-forward, RNN, or CNN architectures, Echo Networks consist solely of a connection matrix, enabling direct manipulation of the network's genome via matrix operations. The authors demonstrate the efficacy of Echo Networks on electrocardiography signal classification and highlight the potential of their matrix-based genome representation for facilitating systematic mutation and recombination within evolutionary algorithms.
Matrix-based genomes in Echo Networks unlock systematic mutation and recombination via matrix operations, potentially revolutionizing neuroevolution for resource-constrained applications.
For applications on the extreme edge, minimal networks of only a few dozen artificial neurons for event detection and classification in discrete time signals would be highly desirable. Feed-forward networks, RNNs, and CNNs evolved through evolutionary algorithms can all be successful in this respect but pose the problem of allowing little systematicity in mutation and recombination if the standard direct genetic encoding of the weights is used (as for instance in the classic NEAT algorithm). We therefore introduce Echo Networks, a type of recurrent network that consists of the connection matrix only, with the source neurons of the synapses represented as rows, destination neurons as columns and weights as entries. There are no layers, and connections between neurons can be bidirectional but are technically all recurrent. Input and output can be arbitrarily assigned to any of the neurons and only use an additional (optional) function in their computational path, e.g., a sigmoid to obtain a binary classification output. We evaluated Echo Networks successfully on the classification of electrocardiography signals but see the most promising potential in their genome representation as a single matrix, allowing matrix computations and factorisations as mutation and recombination operators.