Mismatch responses for activity regularization condition: A. Predictive global; B. Predictive local; C: Contrastive global; D: Contrastive local, E: Supervised; F: Untrained.

Mismatch responses for activity regularization condition: A. Predictive global; B. Predictive local; C: Contrastive global; D: Contrastive local, E: Supervised; F: Untrained.

<p>Each subfigure shows the Mean Squared Average activation of a layer over different time steps, separated for sequences when an unexpected change is happening versus when an expected change is happening. Shaded areas show 95% Confidence intervals.</p>

Saved in:
Bibliographic Details
Main Author: Dirk Gütlin (22566778) (author)
Other Authors: Ryszard Auksztulewicz (4640515) (author)
Published: 2025
Subjects:
Biophysics
Neuroscience
Physiology
Science Policy
Biological Sciences not elsewhere classified
Information Systems not elsewhere classified
without evaluating whether
like effects across
like behaviors better
including mismatch responses
biological neural networks
compared two pc
advancing brain modelling
inspired training objectives
like responses
training algorithms
neural learning
work contributes
underlying mechanisms
semantic information
saving principles
results show
promising foundation
pc framework
neuroscientific theory
key signatures
inspired algorithms
important mechanism
gain control
findings indicate
dnn ).
contrastive approach
accurately reflect
Tags: Add Tag
No Tags, Be the first to tag this record!

Similar Items