Neural Network Concept Testing via Backpropagated CAV Attribution

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current XAI approaches for neural networks are limited by their focus on single neurons or channels, restricting concept space and failing to consider overlaps or human-defined concepts, making it difficult to interpret and verify model decisions, especially in safety-critical applications.

Innovation Solution

The method involves obtaining a concept activation vector (CAV) to process attributions through the neural network using backpropagation, allowing for the detection and localization of concepts across multiple channels, thereby enhancing interpretability and enabling the testing of predefined concepts within the network.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If local attribution methods are used to provide transparency for single model decisions, then interpretability for single inputs is improved, but the concept space is restricted to single neurons or channels

Engineering Contradiction:
ImproveinterpretabilityVSAvoidconcept space
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent combines multiple neurons or channels into a single concept vector that represents a unified concept. Instead of analyzing individual neurons separately, the method merges their contributions into a cohesive concept representation that can be attributed to model decisions, thereby expanding the concept space while maintaining interpretability.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The concept vector serves multiple functions: it represents a semantic concept, aggregates information from multiple neurons/channels, and enables attribution analysis. This multi-functionality allows the same mechanism to handle both local interpretability and broader concept space coverage.

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Measurement precision

If global explainability methods are used to examine model behavior generally, then broader model understanding is improved, but overlap of different concepts in latent space is not considered

Engineering Contradiction:
Improvemodel understandingVSAvoidconcept overlap handling
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent applies local quality by allowing different concept vectors to have distinct characteristics and representations in the latent space. Each concept is treated with its own unique vector representation that captures its specific properties, enabling the model to handle overlapping concepts by differentiating them through their local qualities rather than forcing a single global representation.

Inventive Principle:
Principle #3Local quality

3Ease of operation

If concepts are split into subspaces or channels for analysis, then local attribution is simplified, but semantic overlap between concepts is lost

Engineering Contradiction:
Improveattribution computationVSAvoidsemantic overlap
Core Design Contradiction:
Ease of operationVSLoss of information

Solution Approach 1:

The concept vector acts as a composite representation that integrates information from multiple neurons and channels while preserving semantic relationships. Rather than treating concept components as separate immutable entities, the composite vector allows information to be combined in a way that maintains semantic overlap and relationships between concepts.

Inventive Principle:
Principle #40Composite materials

Data Source

PatentEP4459505A1Apparatus, computer program and method for testing a neural network for a concept
Publication Date: 2024.11.06 CONTINENTAL AUTOMOTIVE TECHNOLOGIES GMBH
  • EP4459505A1 patent drawingFigure 1
  • EP4459505A1 patent drawingFigure 2
  • EP4459505A1 patent drawingFigure 3

AI summary

Embodiments of the present disclosure relate to an apparatus (600), a computer program, and a method (100) for testing a neural network for a concept. The method provides for passing an output of the neural network backwards through the neural network and, in doing so, processing one or more attributions of the concept layer based on the CAV to obtain a projected attribution. As well, the method suggests testing the neural network for the concept using the projected attribution.