Embedded Files
Learn Z0 ≡ color and Z1 ≡ object type. Latent dimension Z0 is passed to the classifier for color and latent dimension Z1 is passed to the classifier for object type. The user-uttered labels for color are red, blue and for object type are juggle ball, orb. All images which can not be described by the user-uttered labels—lego bricks and whiteboard pins, rubber ducks or green objects—are given an unknown ground truth label.
The task the human performs is to separate a set of observed objects by their function - juggling balls vs orbs, and then by their color - red vs yellow vs blue. Lego blocks and whiteboard pins are also present in the scene, but they are not manipulated and no label information is given about them from the expert. At test time the agent has to repeat the task, with new objects being present in the scene that were previously unobserved—green objects and a yellow rubber duck
Full Model (Alpha = 1, Beta = 30, Gamma = 10000)
Full Model (Alpha = 1, Beta = 30, Gamma = 10000)
Scatter plot of the observed validation data points for concept group 0 (color) over Z0 (x axis) and Z1 (y axis)
Scatter plot of the observed validation data points for concept group 1 (object type) over Z0 (x axis) and Z1 (y axis)
Per-concept group confusion matrix for classifying every observed image part of the real-objects dataset. The labels are consistent with the the description at the top of the page. The statistics from the matrices are used to calculate the reported F1 scores in the paper.
Scatter plot of both the observed and unobserved validation data points for concept group 0 (color) over Z0 (x axis) and Z1 (y axis)
Scatter plot both the observed and unobserved validation data points for concept group 1 (object type) over Z0 (x axis) and Z1 (y axis)
Per-concept group confusion matrix for classifying every unobserved image part of the real-objects dataset. The labels are consistent with the the description at the top of the page. The statistics from the matrices are used to calculate the reported F1 scores in the paper.
Cosine similarity diagrams for the labels from concept group 0 (blue, red, yellow) and concept group 1 (juggling ball, orb). The average entropy of the distributions of normalised cosine values for the white cells, for each label, is reported in the paper. The value below each component from the PCA analysis represents the eigen value for that component.
Classifier Network (Alpha = 0, Beta = 30, Gamma = 10000)
Classifier Network (Alpha = 0, Beta = 30, Gamma = 10000)
Scatter plot of the observed validation data points for concept group 0 (color) over Z0 (x axis) and Z1 (y axis)
Scatter plot of the observed validation data points for concept group 1 (object type) over Z0 (x axis) and Z1 (y axis)
Per-concept group confusion matrix for classifying every observed image part of the real-objects dataset. The labels are consistent with the the description at the top of the page. The statistics from the matrices are used to calculate the reported F1 scores in the paper.
Scatter plot of both the observed and unobserved validation data points for concept group 0 (color) over Z0 (x axis) and Z1 (y axis)
Scatter plot of both the observed and unobserved validation data points for concept group 1 (object type) over Z0 (x axis) and Z1 (y axis)
Per-concept group confusion matrix for classifying every unobserved image part of the real-objects dataset. The labels are consistent with the the description at the top of the page. The statistics from the matrices are used to calculate the reported F1 scores in the paper.
Cosine similarity diagrams for the labels from concept group 0 (blue, red, yellow) and concept group 1 (juggling ball, orb). The average entropy of the distributions of normalised cosine values for the white cells, for each label, is reported in the paper. The value below each component from the PCA analysis represents the eigen value for that component.
Beta-VAE (Alpha = 1, Beta = 30, Gamma = 0)
Beta-VAE (Alpha = 1, Beta = 30, Gamma = 0)
Scatter plot of the observed validation data points for concept group 0 (color) over Z0 (x axis) and Z1 (y axis)
Scatter plot of the observed validation data points for concept group 1 (object type) over Z0 (x axis) and Z1 (y axis)
Per-concept group confusion matrix for classifying every observed image part of the real-objects dataset. The labels are consistent with the the description at the top of the page. The statistics from the matrices are used to calculate the reported F1 scores in the paper.
Scatter plot of both the observed and unobserved validation data points for concept group 0 (color) over Z0 (x axis) and Z1 (y axis)
Scatter plot of both the observed and unobserved validation data points for concept group 1 (object type) over Z0 (x axis) and Z1 (y axis)
Per-concept group confusion matrix for classifying every unobserved image part of the real-objects dataset. The labels are consistent with the the description at the top of the page. The statistics from the matrices are used to calculate the reported F1 scores in the paper.
Cosine similarity diagrams for the labels from concept group 0 (blue, red, yellow) and concept group 1 (juggling ball, orb). The average entropy of the distributions of normalised cosine values for the white cells, for each label, is reported in the paper. The value below each component from the PCA analysis represents the eigen value for that component.
Google Sites
Report abuse