concept
active
concept:softmax-competition-for-priority-logicSoftmax competition for priority logic
Heads attend to A if present else B via key scoring.
Neighborhood — ranked by edge-count
Papers (1)
paper
Questions (1)
question
- Open question about competition among keys.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Policies assigned probability via softmax of expected free energy; enables self-evidencing behavior.
- Failure mode for output-surjectivity: LLMs may lack capacity to predict all tokens due to rank constraints
- Selecting policies using a softmax (normalized exponential) function of negative expected free energy.
- Organisation that hosted the Holistic Intelligence unconference where the paper's ideas originated
- Neuronal dynamics computed from free energy gradients; interpreted as average firing rate of neural populations.
- Proximal operator of the l2,1 norm used as activation.
- Neural plausibility argument for softmax policy selection.
- The novel framework introduced in this paper, combining DLGN and NCA for fully differentiable discrete CA learning