PA-Triplet Loss
A triplet loss function that automatically partitions positive and negative sample pairs based on distances in the generation parameter space. It converts generation parameters directly into relational supervisory signals and is used for metric learning.
Timeline (last 90 days)
FDSML (Formula-Driven Supervised Metric Learning) shifts the supervision signal from discrete class labels to image generation parameters, constructing relational supervision through a parameter-aware triplet loss (PA-Triplet Loss).
FDSML consists of two key components: the FineFractalDB dataset and the PA-Triplet Loss.
PA-Triplet Loss assigns positive and negative sample pairs based on distances in the underlying iterated function system (IFS) parameter space: samples close in parameter space are treated as positive pairs, and those far apart as negative pairs.
All Facts (3)
FDSML (Formula-Driven Supervised Metric Learning) shifts the supervision signal from discrete class labels to image generation parameters, constructing relational supervision through a parameter-aware triplet loss (PA-Triplet Loss).
50%UnverifiedFDSML consists of two key components: the FineFractalDB dataset and the PA-Triplet Loss.
50%UnverifiedPA-Triplet Loss assigns positive and negative sample pairs based on distances in the underlying iterated function system (IFS) parameter space: samples close in parameter space are treated as positive pairs, and those far apart as negative pairs.
50%