Feature (neural network)

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Features are the fundamental unit of neural networks; they correspond to directions and can be rigorously studied and understood.claim0.858
  First of three speculative claims forming the foundation of the circuits research agenda
• Neural-network agentsconcept0.799
  The artificial agents trained with RL in this study, whose latent dynamics are analyzed for causal emergence.
• Convolutional Neural Networksmethod0.789
  Biologically-inspired AI architecture cited as a successful example of bioinspiration from visual cortex organization
• neural cognitionconcept0.779
  Cognition in nervous systems, used as a modelling target
• Neural codeconcept0.777
  The model's parameters considered as the actual 'code' implementing its algorithms, as opposed to human-written code.
• Feature engineeringconcept0.777
  Domain of techniques for constructing informative features from raw data; covariance pooling is a feature engineering method for token sequences.
• Feature Neurons and Memory Neurons Architectureframework0.768
  Biologically plausible two-pool architecture from Krotov & Hopfield (2020) splitting self-attention into feature and memory neuron populations; used to interpret TEM-t place cells.
• Neural Network Interpretabilityconcept0.767
  The field aimed at understanding what neural networks have learned; characterized as pre-paradigmatic in this paper