# Martijho-PathNet-search-plots

(Forskjeller mellom versjoner)
Gå til: navigasjon, søk
 Versjonen fra 28. feb 2018 kl. 14:39 (vis kilde) (→Diversity metrics)← Eldre redigering Versjonen fra 28. feb 2018 kl. 14:40 (vis kilde) (→Diversity metrics)Nyere redigering → Linje 3: Linje 3: As a calculation of diversity within one population, an average euclidean distance to a population centroid was used for each layer and the averaged. As a calculation of diversity within one population, an average euclidean distance to a population centroid was used for each layer and the averaged. - \frac{1 N} \sum{i=1}^N X_{i}[/itex] + \frac{1 N} \sum{i=1}^N Xi [/itex] === Home-made for each layer === === Home-made for each layer ===

## Population Diversity

### Diversity metrics

As a calculation of diversity within one population, an average euclidean distance to a population centroid was used for each layer and the averaged.

Feil i matematikken (Mangler kjørbar texvc; se math/README for oppsett.): \frac{1 N} \sum{i=1}^N Xi

### Home-made for each layer  Average population diversity for each layer in generation. Experiment: Recomb  Average population diversity for each layer in generation. Experiment: Low  Average population diversity for each layer in generation. Experiment: High  Average population diversity for each layer in generation. Experiment: Low to High  Average population diversity for each layer in generation. Experiment: High to Low

## Path size  The change in average path size for each generation.  The average size of each layer in each optimal path found.

## Capacity and Reuse  Average number of modules used during the different searches plotted consecutively. The gray dotted line is what number of modules would be used if modules were selected randomly for each task.  Used capacity in a PathNet plotted against that networks average classification accuracy for all tasks  Average reuse for each generation plotted for each experiment type. The gray dotted line is what reuse would be reached by random module selection

## Training  Average training accuracy for each of the five search algorithms during the search for each task. The dotted line is the validation accuracy reached after the search was completed.  Number of training iterations in total for all modules in the PathNet plotted against the cumulative validation accuracy reached for all tasks. The size of the circle corresponds to the amount of modules used.  Plot of the ratio of unit training in locked and stored modules on the total number of unit training in all modules for each task