Modular and Non-Modular Neural Network Efficiency Simulations

Simulation data analyzes the functional efficiency of modular versus non-modular neural networks across different sizes. The study includes detailed comparisons for 'small' and 'large' network size classes and a broader analysis across a range of network sizes. It was authored by Colin R. Tosh and published in 2020.

Use Cases

• Compare efficiency metrics between modular and non-modular network architectures across defined size classes.
• Analyze how relative functional performance scales with network size for different structural types.
• Investigate the relationship between low connective density and network efficiency, a factor mentioned as relevant for energy constraints.

Strengths

• Analysis is grounded in a comparative framework of 'small' and 'large' network size classes.
• Dataset supports investigation into the evolutionary relationship between network size and modularity prevalence.
• Published under a permissive CC0 1.0 license for unrestricted reuse.

Limitations

• Specific data structure, column names, row counts, and file formats are not provided in the input.
• The dataset's creation method and the exact metrics of 'efficiency' are not detailed.
• Data is from 2020 and may not reflect the latest computational models or findings in network theory.

Provenance

Source

dryad

Collection Method

Simulation data from a computational study comparing network architectures.

Freshness

Last updated on 2020-06-30.