Back to search

Article

Shared architecture, competing objectives: experimental evolution in Drosophila instantiates negative transfer in multi-task learning

2026-06-22

Abstract excerpt

<title>Abstract</title> <p> When competing objectives are optimized through shared components, gains on one can degrade performance on another. Evolutionary biologists call this a fitness trade-off; machine learning researchers call it negative transfer. The structural parallel is rarely made explicit, and the two literatures have developed independently despite describing the same underlying constraint. Here I...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work
f6157599-c238-580c-ae00-ef4c5a91ce28
DOI
10.21203/rs.3.rs-10084767/v1
Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 11 closest
Interactive article relationship graphSelect a related publication card to move it into the centre and load its closest explainable connections. Solid lines are source-backed structured connections. Dashed lines are semantic discovery signals and are not scientific evidence.
Shared architecture, competing objectives: experimental evolution in Drosophila instantiates negative transfer in multi-task learningDOI 10.21203/rs.3.rs-10084767/v1
Select a neighboring publication to make it the new centre.