Publicação
On The Topological Expressive Power of Neural Networks
| Resumo: | We propose a topological description of neural network expressive power. We adopt the topology of the space of decision boundaries realized by a neural architecture as a measure of its intrinsic expressive power. By sampling a large number of neural architectures with different sizes and design, we show how such measure of expressive power depends on the properties of the architectures, like depth, width and other related quantities. |
|---|---|
| Autores principais: | Petri, Giovanni |
| Outros Autores: | Leitão, António |
| Assunto: | Neural Networks Expressive Power Decision Boundary Classification |
| Ano: | 2020 |
| País: | Portugal |
| Tipo de documento: | póster em conferência |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade Nova de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório Institucional da UNL |
| Resumo: | We propose a topological description of neural network expressive power. We adopt the topology of the space of decision boundaries realized by a neural architecture as a measure of its intrinsic expressive power. By sampling a large number of neural architectures with different sizes and design, we show how such measure of expressive power depends on the properties of the architectures, like depth, width and other related quantities. |
|---|
Atividades financiadas
Não foram encontrados projetos financiados.