Reducing Overconfidence Predictions for Autonomous Driving Perception

TL;DR

This paper proposes a probabilistic approach using Logit layer scores and ML/MAP functions to reduce overconfidence in deep learning-based object recognition for autonomous driving, improving interpretability without retraining.

Contribution

It introduces a method that replaces SoftMax and Sigmoid with ML and MAP functions on pre-trained networks for better probabilistic predictions in autonomous perception.

Findings

ML and MAP outperform SoftMax and Sigmoid in probabilistic accuracy

Approach works with RGB and LiDAR data from KITTI and Lyft datasets

No retraining needed, only modification during inference

Abstract

In state-of-the-art deep learning for object recognition, SoftMax and Sigmoid functions are most commonly employed as the predictor outputs. Such layers often produce overconfident predictions rather than proper probabilistic scores, which can thus harm the decision-making of `critical' perception systems applied in autonomous driving and robotics. Given this, the experiments in this work propose a probabilistic approach based on distributions calculated out of the Logit layer scores of pre-trained networks. We demonstrate that Maximum Likelihood (ML) and Maximum a-Posteriori (MAP) functions are more suitable for probabilistic interpretations than SoftMax and Sigmoid-based predictions for object recognition. We explore distinct sensor modalities via RGB images and LiDARs (RV: range-view) data from the KITTI and Lyft Level-5 datasets, where our approach shows promising performance compared to the usual SoftMax and Sigmoid layers, with the benefit of enabling interpretable probabilistic predictions. Another advantage of the approach introduced in this paper is that the ML and MAP functions can be implemented in existing trained networks, that is, the approach benefits from the output of the Logit layer of pre-trained networks. Thus, there is no need to carry out a new training phase since the ML and MAP functions are used in the test/prediction phase.