English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep Learning (DL) is a surprisingly successful branch of machine learning.The success of DL is usually explained by focusing analysis on a particularrecent algorithm and its traits. Instead, we propose that an explanation of thesuccess of DL must look at the population of all algorithms in the field andhow they have evolved over time. We argue that cultural evolution is a usefulframework to explain the success of DL. In analogy to biology, we use`development' to mean the process converting the pseudocode or text descriptionof an algorithm into a fully trained model. This includes writing theprogramming code, compiling and running the program, and training the model. Ifall parts of the process don't align well then the resultant model will beuseless (if the code runs at all!). This is a constraint. A core component ofevolutionary developmental biology is the concept of deconstraints -- these aremodification to the developmental process that avoid complete failure byautomatically accommodating changes in other components. We suggest that manyimportant innovations in DL, from neural networks themselves to hyperparameteroptimization and AutoGrad, can be seen as developmental deconstraints. Thesedeconstraints can be very helpful to both the particular algorithm in how ithandles challenges in implementation and the overall field of DL in how easy itis for new ideas to be generated. We highlight how our perspective can bothadvance DL and lead to new insights for evolutionary biology.

Nothing makes sense in deep learning, except in the light of evolution