An improved fixation point generator for TDFN is trained to match the difference between the network's reconstructed image and the input image, yielding better MNIST accuracy with fewer fixations than the prior RL approach.
Decision- Theoretic Saliency : Computational Principles , Biological Plausibility , and Implications for Neurophysiology and Psychophysics
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Advancing TDFN: Precise Fixation Point Generation Using Reconstruction Differences
An improved fixation point generator for TDFN is trained to match the difference between the network's reconstructed image and the input image, yielding better MNIST accuracy with fewer fixations than the prior RL approach.