Towards Structured Evaluation of Deep Neural Network Supervisors

• Deep Neural Networks (DNN) have improved the quality of several non-safety related products in the past years. However, before DNNs should be deployed to safety-critical applications, their robustness needs to be systematically analyzed. A common challenge for DNNs occurs when input is dissimilar to the training set, which might lead to high confidence predictions despite proper knowledge of the input. Several previous studies have proposed to complement DNNs with a supervisor that detects when inputs are outside the scope of the network. Most of these supervisors, however, are developed and tested for a selected scenario using a specific performance metric. In this work, we emphasize the need to assess and compare the performance of supervisors in a structured way. We present a framework constituted by four datasets organized in six test cases combined with seven evaluation metrics. The test cases provide varying complexity and include data from publicly available sources as well as a novel dataset consisting of images from simulated driving scenarios. The latter we plan to make publicly available. Our framework can be used to support DNN supervisor evaluation, which in turn could be used to motive development, validation, and deployment of DNNs in safety-critical applications.

Subject headings

TEKNIK OCH TEKNOLOGIER  -- Annan teknik (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Other Engineering and Technologies (hsv//eng)

NATURVETENSKAP  -- Data- och informationsvetenskap -- Programvaruteknik (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Software Engineering (hsv//eng)

TEKNIK OCH TEKNOLOGIER  -- Elektroteknik och elektronik -- Datorsystem (hsv//swe)

ENGINEERING AND TECHNOLOGY  -- Electrical Engineering, Electronic Engineering, Information Engineering -- Computer Systems (hsv//eng)

NATURVETENSKAP  -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)

Keyword

Automotive perception

Deep neural networks

Out-of-distribution

Robustness

Supervisor

Artificial intelligence

Robustness (control systems)

Safety engineering

Statistical tests

Supervisory personnel

Evaluation metrics

High confidence

Performance metrices

Safety critical applications

Safety-related products

Simulated driving

Structured evaluation

deep neural networks

robustness

out-of-distribution

supervisor

automotive perception

OCEEDINGS7th Industrial Conference on Data Mining

JUL 14-18

2007

Leipzig

GERMANY

V4597

OCESSING (WCSP)7th International Conference on Wireless Communications and Signal Processing

Publication and Content Type

ref (subject category)

kon (subject category)