Upgrade to Pro — share decks privately, control downloads, hide ads and more …

DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Cars

DeepTest: Automated Testing of Deep-Neural-Network-driven Autonomous Cars

Presentation of a paper by Tian et al. (2017) at the Mila Robotics RG: https://arxiv.org/abs/1708.08559

When: Monday, 10/08/18 at 2.00pm
Where: PAA 3195

Synthesizing images through deformation or generation has a long history in computer vision and adversarial machine learning. More broadly, this technique is also known as data augmentation [1] [2], feature augmentation [3], or domain randomization [2] (in the non-adversarial setting). Often used to augment a dataset to increase data diversity, to probe a model for hidden biases, or reduce sensitivity to noise, this family of techniques seek to generate realistic but synthetic inputs based on our understanding of the data generating process, and in some cases, the model architecture. Such inputs can be used to discover hidden failure modes and improve generalization through retraining.

In this work, Tian et al. present a grey-box testing technique to evaluate deep neural networks by applying realistic image deformations to maximize “neuron coverage” and promote output diversity. This technique is used to discover images that would induce unsafe control outputs in a self driving car, i.e. dangerous steering angles. In this talk, we will introduce the concept “neuron coverage” and its usefulness as a proxy for capturing output diversity. We will explore the notion of metamorphic testing, and related white-box techniques for evaluating neural networks. And we will learn how fake training data can help avoid rapid unplanned deceleration of an autonomous vehicle.

[1] Understanding data augmentation for classification: When to warp? https://arxiv.org/abs/1609.08764
[2] The Effectiveness of Data Augmentation in Image Classification using Deep Learning https://arxiv.org/abs/1712.04621
[3] Dataset augmentation in feature space https://arxiv.org/abs/1702.05538
[4] Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World https://arxiv.org/abs/1703.06907

Breandan Considine

July 18, 2023
Tweet

More Decks by Breandan Considine

Other Decks in Research

Transcript

  1. DeepTest:
    Automated Testing of
    Deep-Neural-Network-driven
    Autonomous Cars
    Yuchi Tian, Kexin Pei, Suman Jana, Baishakhi Ray
    Presentation by Breandan Considine

    View full-size slide

  2. How do you test an
    arbitrary program?
    P
    x1 P(x )
    1

    View full-size slide

  3. P
    x1
    P(x )
    1
    O(x )
    1
    =

    O

    View full-size slide

  4. P P(x )
    2
    O(x )
    2
    =

    x2
    O

    View full-size slide

  5. P P(x )
    3
    O(x )
    3
    =

    x3
    O

    View full-size slide

  6. O
    P P(x )
    3
    O(x )
    3
    =

    x3

    View full-size slide

  7. Metamorphic Relations
    P
    P(g(x ))
    i
    xi
    P
    g(x )
    i
    P(x )
    i
    =, =, ⊂, ⊊, ...

    View full-size slide

  8. Metamorphic Relations
    S(A, B) : Returns shortest path from A to B

    View full-size slide

  9. What are some necessary properties off?
    f(A, B) = f(B, A)
    x ∈ S(A, B), S(A, x) + S(x, B) = S(A, B)
    z ∈ G ∖ S(A, B), ∣S(A, B)∣ ≤ ∣S(A, z) + S(z, B)∣
    x ∈ S(A, B), ∣S(A, x)∣ < ∣S(A, B)∣
    x ∈ S(A, B), S(A, x) ⊂ S(A, B)

    View full-size slide

  10. Metamorphic Relations
    S
    S(B, A)
    A, B
    S
    B, A
    S(A, B)
    =

    View full-size slide

  11. Metamorphic Relations
    S
    S(B, A)
    A, B
    S
    B, A
    S(A, B)
    =

    View full-size slide

  12. Neuron Coverage
    ∣total neurons∣
    ∣activated neurons∣

    View full-size slide

  13. ( − θ ) ≤ λMSE
    θi
    ^
    ti
    2
    origin
    Metamorphic Relation

    View full-size slide

  14. N
    ( − N(T(i)))
    θi
    ^ 2
    i
    N
    T(i)
    <
    ( − N(i))
    θi
    ^ 2
    Metamorphic Relation

    View full-size slide