Evaluating the Impact of Uncertainty Visualization on Model Reliance

Zhao, J. L.; Wang, Yixuan; Mancenido, Michelle V.; Chiou, Erin K.; Maciejewski, Ross

doi:10.1109/tvcg.2023.3251950

Cited by 4 publications

(12 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Key concepts in this context are trust and reliance [39,80,86]. Trust reflects subjective perceptions of AI, often assessed using self-report scales, while reliance pertains to objective behavior in response to AI systems [2,47,84,90,91]. Intriguingly, studies have revealed inconsistencies between trust and reliance.…”

Section: Related Work 21 Appropriate Reliance In Ai-assisted Decision...mentioning

confidence: 99%

“…Behavior-Based Measurement: This approach assesses appropriate reliance by analyzing human behaviors, considering AI confidence as an indicator of trustworthiness [90,91]. If humans rely more on AI's suggestions when the AI expresses high confidence (and rely on AI less when AI confidence is low), it's deemed calibrated and appropriate.…”

Section: Related Work 21 Appropriate Reliance In Ai-assisted Decision...mentioning

confidence: 99%

“…However, the confidence of AI may not directly represent the correctness of AI. Therefore, many works that rely on this measure of the appropriateness of human reliance have found that although people's trust/reliance gets calibrated, the final task performance does not improve [90,91].…”

Section: Related Work 21 Appropriate Reliance In Ai-assisted Decision...mentioning

confidence: 99%

“…For instance, Rastogi et al [67] adjusted decision-making timeframes based on AI confidence levels to promote humans' more analytical thinking when AI's confidence is low. Besides, various confidence representation methods, including violin plots or question marks, have also been explored [91]. However, these approaches haven't consistently resulted in improved reliance appropriateness or task performance [47,90,91] -"only displaying AI confidence can be insufficient".…”

Section: Enhancing Appropriate Reliance In Ai-assisted Decision-makingmentioning

confidence: 99%

“…Given the inherent uncertainties of both humans and AI, one key challenge is ensuring humans' appropriate reliance on AI [2,90]. Showing AI confidence levels has been proposed to address this, as accurate confidence scores can indicate the likelihood of correct predictions [2,38,47,67,90,91]. Nonetheless, studies on AI confidence presentation show mixed outcomes, suggesting it doesn't always improve human-AI collaboration outcomes [47,90,91].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Will You Accept the AI Recommendation? Predicting Human Behavior in AI-Assisted Decision Making

Wang

Yin

2022

Proceedings of the ACM Web Conference 2022

View full text Add to dashboard Cite

In AI-assisted decision-making, it is crucial but challenging for humans to achieve appropriate reliance on AI. This paper approaches this problem from a human-centered perspective, "human selfconfidence calibration". We begin by proposing an analytical framework to highlight the importance of calibrated human self-confidence. In our first study, we explore the relationship between human selfconfidence appropriateness and reliance appropriateness. Then in our second study, We propose three calibration mechanisms and compare their effects on humans' self-confidence and user experience. Subsequently, our third study investigates the effects of self-confidence calibration on AI-assisted decision-making. Results show that calibrating human self-confidence enhances human-AI team performance and encourages more rational reliance on AI (in some aspects) compared to uncalibrated baselines. Finally, we discuss our main findings and provide implications for designing future AI-assisted decision-making interfaces. CCS CONCEPTS• Human-centered computing → Empirical studies in HCI.

show abstract

Section: Related Work 21 Appropriate Reliance In Ai-assisted Decision...mentioning

confidence: 99%

Section: Related Work 21 Appropriate Reliance In Ai-assisted Decision...mentioning

confidence: 99%

Section: Related Work 21 Appropriate Reliance In Ai-assisted Decision...mentioning

confidence: 99%

Section: Enhancing Appropriate Reliance In Ai-assisted Decision-makingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Will You Accept the AI Recommendation? Predicting Human Behavior in AI-Assisted Decision Making

Wang

Yin

2022

Proceedings of the ACM Web Conference 2022

View full text Add to dashboard Cite

show abstract

Uncertainty quantification for probabilistic machine learning in earth observation using conformal prediction

Singh,

Moncrieff,

Venter

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Machine learning is increasingly applied to Earth Observation (EO) data to obtain datasets that contribute towards international accords. However, these datasets contain inherent uncertainty that needs to be quantified reliably to avoid negative consequences. In response to the increased need to report uncertainty, we bring attention to the promise of conformal prediction within the domain of EO. Unlike previous uncertainty quantification methods, conformal prediction offers statistically valid prediction regions while concurrently supporting any machine learning model and data distribution. To support the need for conformal prediction, we reviewed EO datasets and found that only 22.5% of the datasets incorporated a degree of uncertainty information, with unreliable methods prevalent. Current open implementations require moving large amounts of EO data to the algorithms. We introduced Google Earth Engine native modules that bring conformal prediction to the data and compute, facilitating the integration of uncertainty quantification into existing traditional and deep learning modelling workflows. To demonstrate the versatility and scalability of these tools we apply them to valued EO applications spanning local to global extents, regression, and classification tasks. Subsequently, we discuss the opportunities arising from the use of conformal prediction in EO. We anticipate that accessible and easy-to-use tools, such as those provided here, will drive wider adoption of rigorous uncertainty quantification in EO, thereby enhancing the reliability of downstream uses such as operational monitoring and decision-making.

show abstract

“Can You Handle the Truth?”: Investigating the Effects of AR-Based Visualization of the Uncertainty of Deep Learning Models on Users of Autonomous Vehicles

Doula,

Schmidt,

Mühlhäuser

et al. 2023

2023 IEEE International Symposium on Mixed and Augmented Reality (ISMAR)

View full text Add to dashboard Cite

Evaluating the Impact of Uncertainty Visualization on Model Reliance

Cited by 4 publications

References 63 publications

Will You Accept the AI Recommendation? Predicting Human Behavior in AI-Assisted Decision Making

Will You Accept the AI Recommendation? Predicting Human Behavior in AI-Assisted Decision Making

Uncertainty quantification for probabilistic machine learning in earth observation using conformal prediction

“Can You Handle the Truth?”: Investigating the Effects of AR-Based Visualization of the Uncertainty of Deep Learning Models on Users of Autonomous Vehicles

Contact Info

Product

Resources

About