Method for understanding machine-learning decisions based on camera data
First Claim
1. A system for understanding machine-learning (ML) decisions, the system comprising:
- one or more processors and a non-transitory computer-readable medium having executable instructions encoded thereon such that when executed, the one or more processors perform operations of;
processing a set of input data with a convolutional neural network (CNN) model, resulting in a set of latent variable activation vectors within the CNN model, wherein each latent variable activation vector in the CNN model represents a concept in the set of input data;
clustering the set of latent variable activation vectors in an unsupervised manner, resulting in a plurality of clusters of latent variable activation vectors;
computing functional semantics for each cluster, wherein the functional semantics represent relationships among concepts in the set of input data;
generating a concept network comprising a plurality of nodes and a plurality of weighted directed edges, wherein the plurality of nodes are defined by concepts and the plurality of weighted directed edges are defined by the computed functional semantics;
in an operational phase, generating a subnetwork of the concept network; and
displaying nodes of the subnetwork as a set of visual images that are annotated by weights and labels; and
refining the CNN model per the weights and labels.
1 Assignment
0 Petitions
Accused Products
Abstract
Described is a system for understanding machine-learning decisions. In an unsupervised learning phase, the system extracts, from input data, concepts represented by a machine-learning (ML) model in an unsupervised manner by clustering patterns of activity of latent variables of the concepts, where the latent variables are hidden variables of the ML model. The extracted concepts are organized into a concept network by learning functional semantics among the extracted concepts. In an operational phase, a subnetwork of the concept network is generated. Nodes of the subnetwork are displayed as a set of visual images that are annotated by weights and labels, and the ML model per the weights and labels.
5 Citations
17 Claims
-
1. A system for understanding machine-learning (ML) decisions, the system comprising:
-
one or more processors and a non-transitory computer-readable medium having executable instructions encoded thereon such that when executed, the one or more processors perform operations of; processing a set of input data with a convolutional neural network (CNN) model, resulting in a set of latent variable activation vectors within the CNN model, wherein each latent variable activation vector in the CNN model represents a concept in the set of input data; clustering the set of latent variable activation vectors in an unsupervised manner, resulting in a plurality of clusters of latent variable activation vectors; computing functional semantics for each cluster, wherein the functional semantics represent relationships among concepts in the set of input data; generating a concept network comprising a plurality of nodes and a plurality of weighted directed edges, wherein the plurality of nodes are defined by concepts and the plurality of weighted directed edges are defined by the computed functional semantics; in an operational phase, generating a subnetwork of the concept network; and displaying nodes of the subnetwork as a set of visual images that are annotated by weights and labels; and refining the CNN model per the weights and labels. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 17)
-
-
9. A computer implemented method for understanding machine-learning (ML) decisions, the method comprising an act of:
-
causing one or more processers to execute instructions encoded on a non-transitory computer-readable medium, such that upon execution, the one or more processors perform operations of; processing a set of input data with a convolutional neural network (CNN) model, resulting in a set of latent variable activation vectors within the CNN model, wherein each latent variable activation vector in the CNN model represents a concept in the set of input data; clustering the set of latent variable activation vectors in an unsupervised manner, resulting in a plurality of clusters of latent variable activation vectors; computing functional semantics for each cluster, wherein the functional semantics represent relationships among concepts in the set of input data; generating a concept network comprising a plurality of nodes and a plurality of weighted directed edges, wherein the plurality of nodes are defined by concepts and the plurality of weighted directed edges are defined by the computed functional semantics; in an operational phase, generating a subnetwork of the concept network; displaying nodes of the subnetwork as a set of visual images that are annotated by weights and labels; and refining the CNN model per the weights and labels. - View Dependent Claims (10, 11, 12)
-
-
13. A computer program product for understanding machine-learning (ML) decisions, the computer program product comprising:
computer-readable instructions stored on a non-transitory computer-readable medium that are executable by a computer having one or more processors for causing the processor to perform operations of; processing a set of input data with a convolutional neural network (CNN) model, resulting in a set of latent variable activation vectors within the CNN model, wherein each latent variable activation vector in the CNN model represents a concept in the set of input data; clustering the set of latent variable activation vectors in an unsupervised manner, resulting in a plurality of clusters of latent variable activation vectors; computing functional semantics for each cluster, wherein the functional semantics represent relationships among concepts in the set of input data; generating a concept network comprising a plurality of nodes and a plurality of weighted directed edges, wherein the plurality of nodes are defined by concepts and the plurality of weighted directed edges are defined by the computed functional semantics; in an operational phase, generating a subnetwork of the concept network; displaying nodes of the subnetwork as a set of visual images that are annotated by weights and labels; and refining the CNN model per the weights and labels. - View Dependent Claims (14, 15, 16)
Specification