My student Anne Linja presented her poster “What the AI Saw: Examining human predictions of deep image classification errors” at the Midwest Cognitive Science Conference in Columbus OH May 28, 2019. A link to the poster is here:
Abstract
Deep Image classifiers have made amazing advances in both basic and applied problems in recent years. Nevertheless, they are still very limited and can be foiled by even simple image distortions. Importantly, the way they fail is often unexpected, and sometimes difficult to even understand. Thus, advances in image classifiers made to improve their transparency and the predictability of their errors may make more of a difference than algorithmic improvements that reduce error rates on benchmarks. To understand the types of expectations humans may have, we conducted a study in which students were asked to predict whether a generic AI system would correctly identify 10 classes of tools (axe, hammer, wrench, flashlight, pliers, saw, scissors, screwdriver, tape measure), with a variety of image transforms (e.g., borders, outline filters, additional objects inserted into image, etc.), and also examined how five commercial deep image classifiers performed on the same imagery. Results revealed that humans tended to predict that distortions and distractions would lead to impairment of the AI systems, and although AI failures did incorporate these factors, they also involved many class-level errors (e.g., calling a wrench a tool or a product), and feature-errors (calling a hammer 'metal' or 'wood') not identified by human novice users. Results will be discussed in the context of Explainable AI systems.
Links
BibTeX (Download)
@conference{Linja2019, title = {What the AI Saw: Examining human predictions of deep image classification errors}, author = {Anne Linja and Lamia Alam and Shane T. Mueller}, url = {https://www.researchgate.net/publication/333246714_What_the_AI_saw_Examining_human_predictions_of_deep_image_classification_errors }, year = {2019}, date = {2019-05-25}, booktitle = {Midwest Cognitive Science}, address = {Columbus, OH}, abstract = {Deep Image classifiers have made amazing advances in both basic and applied problems in recent years. Nevertheless, they are still very limited and can be foiled by even simple image distortions. Importantly, the way they fail is often unexpected, and sometimes difficult to even understand. Thus, advances in image classifiers made to improve their transparency and the predictability of their errors may make more of a difference than algorithmic improvements that reduce error rates on benchmarks. To understand the types of expectations humans may have, we conducted a study in which students were asked to predict whether a generic AI system would correctly identify 10 classes of tools (axe, hammer, wrench, flashlight, pliers, saw, scissors, screwdriver, tape measure), with a variety of image transforms (e.g., borders, outline filters, additional objects inserted into image, etc.), and also examined how five commercial deep image classifiers performed on the same imagery. Results revealed that humans tended to predict that distortions and distractions would lead to impairment of the AI systems, and although AI failures did incorporate these factors, they also involved many class-level errors (e.g., calling a wrench a tool or a product), and feature-errors (calling a hammer 'metal' or 'wood') not identified by human novice users. Results will be discussed in the context of Explainable AI systems.}, keywords = {XAI}, pubstate = {published}, tppubtype = {conference} }
188 comments on “What the AI Saw”
Comments are closed.