Run-Time Monitoring of Machine Learning for Robotic Perception: A Survey of Emerging Trends
arXiv:2101.01364 · doi:10.1109/ACCESS.2021.3055015
Abstract
As deep learning continues to dominate all state-of-the-art computer vision tasks, it is increasingly becoming an essential building block for robotic perception. This raises important questions concerning the safety and reliability of learning-based perception systems. There is an established field that studies safety certification and convergence guarantees of complex software systems at design-time. However, the unknown future deployment environments of an autonomous system and the complexity of learning-based perception make the generalization of design-time verification to run-time problematic. In the face of this challenge, more attention is starting to focus on run-time monitoring of performance and reliability of perception systems with several trends emerging in the literature. This paper attempts to identify these trends and summarise the various approaches to the topic.
arXiv commentUpdated version of 10.1109/ACCESS.2021.3055015. Published at IEEE Access. 27 January 2021
References in corpus (26)
- What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
- A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges
- Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
- A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
- Can You Trust Your Model's Uncertainty? Evaluating Predictive Uncertainty Under Dataset Shift
- A tutorial on conformal prediction
- Do ImageNet Classifiers Generalize to ImageNet?
- A Review and Comparative Study on Probabilistic Object Detection in Autonomous Driving
- Monocular Depth Estimation Based On Deep Learning: An Overview
- Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift
- Selective Classification for Deep Neural Networks
- Peduncle Detection of Sweet Pepper for Autonomous Crop Harvesting - Combined Colour and 3D Information
- Addressing Failure Prediction by Learning Model Confidence
- Automated Evaluation of Semantic Segmentation Robustness for Autonomous Driving
- Failing to Learn: Autonomously Identifying Perception Failures for Self-driving Cars
- Evaluating Uncertainty Quantification in End-to-End Autonomous Driving Control
- Metric Learning for Novelty and Anomaly Detection
- Did You Miss the Sign? A False Negative Alarm System for Traffic Sign Detectors
- Deep Verifier Networks: Verification of Deep Discriminative Models with Deep Generative Models
- Time-Dynamic Estimates of the Reliability of Deep Semantic Segmentation Networks
- Outside the Box: Abstraction-Based Monitoring of Neural Networks
- A Rapidly Deployable Classification System using Visual Data for the Application of Precision Weed Management
- Automatic Open-World Reliability Assessment
- Predicting Model Failure using Saliency Maps in Autonomous Driving Systems
- Place Categorization and Semantic Mapping on a Mobile Robot
- Increasing Trustworthiness of Deep Neural Networks via Accuracy Monitoring