LGCVMAApr 26, 2024

On the Road to Clarity: Exploring Explainable AI for World Models in a Driver Assistance System

IBM
arXiv:2404.17350v11 citationsh-index: 20CAI
Originality Incremental advance
AI Analysis

This addresses the need for transparency and safety in autonomous driving systems, though it is incremental as it refines existing XAI techniques for specific models.

The paper tackled the problem of black-box neural networks in autonomous driving by developing explainable AI methods for world models, achieving performance comparable to black-box models with a transparent backbone for VAEs and proposing visualization techniques to analyze internal causes of poor reconstruction.

In Autonomous Driving (AD) transparency and safety are paramount, as mistakes are costly. However, neural networks used in AD systems are generally considered black boxes. As a countermeasure, we have methods of explainable AI (XAI), such as feature relevance estimation and dimensionality reduction. Coarse graining techniques can also help reduce dimensionality and find interpretable global patterns. A specific coarse graining method is Renormalization Groups from statistical physics. It has previously been applied to Restricted Boltzmann Machines (RBMs) to interpret unsupervised learning. We refine this technique by building a transparent backbone model for convolutional variational autoencoders (VAE) that allows mapping latent values to input features and has performance comparable to trained black box VAEs. Moreover, we propose a custom feature map visualization technique to analyze the internal convolutional layers in the VAE to explain internal causes of poor reconstruction that may lead to dangerous traffic scenarios in AD applications. In a second key contribution, we propose explanation and evaluation techniques for the internal dynamics and feature relevance of prediction networks. We test a long short-term memory (LSTM) network in the computer vision domain to evaluate the predictability and in future applications potentially safety of prediction models. We showcase our methods by analyzing a VAE-LSTM world model that predicts pedestrian perception in an urban traffic situation.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes