Neural-network agents trained in social environments learn hybrid navigation strategies that combine individual landmark use with social following, with strategy shifts driven by the ratio of skilled to unskilled social agents.
Optimization and variability can coexist
2 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
2
Pith papers citing it
1
external citations · Pith
abstract
Many biological systems perform close to their physical limits, but promoting this optimality to a general principle seems to require implausibly fine tuning of parameters. Using examples from a wide range of systems, we show that this intuition is wrong. Near an optimum, functional performance depends on parameters in a "sloppy'' way, with some combinations of parameters being only weakly constrained. Absent any other constraints, this predicts that we should observe widely varying parameters, and we make this precise: the entropy in parameter space can be extensive even if performance on average is very close to optimal. This removes a major objection to optimization as a general principle, and rationalizes the observed variability.
years
2026 2verdicts
CONDITIONAL 2representative citing papers
A coarse-grained Levy-jump model predicts that recurrent networks settle into either a collapsed (exponential-forgetting) or anti-collapsed (power-law-forgetting) regime, with one spectral exponent beta governing both the time-scale spectrum and the envelope.
citing papers explorer
-
Social-spatial dependencies for learning visual navigation
Neural-network agents trained in social environments learn hybrid navigation strategies that combine individual landmark use with social following, with strategy shifts driven by the ratio of skilled to unskilled social agents.
-
Anti-Collapse Dynamics and the Emergence of Multi-Time-Scale Learning in Recurrent Neural Networks
A coarse-grained Levy-jump model predicts that recurrent networks settle into either a collapsed (exponential-forgetting) or anti-collapsed (power-law-forgetting) regime, with one spectral exponent beta governing both the time-scale spectrum and the envelope.