No Explainability without Accountability

Smith-Renner, Alison; Ron, Fan; Birchfield, Melissa; Wu, Tongshuang; Boyd-Graber, Jordan; Weld, Daniel S.; Findlater, Leah

doi:10.1145/3313831.3376624

Cited by 81 publications

(22 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While machine learning measures mainly focus on system performance, human-centered machine learning (HCML) investigates how the users are affected by such systems [22]. Accordingly, research in HCML focuses on users' evaluations of machine learning systems, including aspects such as users' reliance [31] on the system, or trust [40,68] in the system, the perceived explainability [51], interpretability [44], or fairness [20,32] of the system, and the experienced or perceived accuracy of the system [15,19,23,54]. In a recent HCML survey paper, Kaluarachchi et al [22] provide a comprehensive overview of user studies with machine learning systems.…”

Section: Human-centered Evaluation Of Machine Learning Systemsmentioning

confidence: 99%

“…While in-person user studies in labs (like in [66]) are rather rare, crowdsourcing tasks are commonly used to collect feedback from the users. Common tasks include labeling text or image data (e.g., in [31]) or assessing given classifications (e.g., in [27,40,48,51]). Often, the classification accuracy is systematically varied to avoid unpredictable behavior by the system [31,40,66].…”

Section: Human-centered Evaluation Of Machine Learning Systemsmentioning

confidence: 99%

See 1 more Smart Citation

How Accurate Does It Feel? – Human Perception of Different Types of Classification Mistakes

Papenmeier

Kern

Hienert

et al. 2022

CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

Supervised machine learning utilizes large datasets, often with ground truth labels annotated by humans. While some data points are easy to classify, others are hard to classify, which reduces the inter-annotator agreement. This causes noise for the classifier and might affect the user's perception of the classifier's performance. In our research, we investigated whether the classification difficulty of a data point influences how strongly a prediction mistake reduces the "perceived accuracy". In an experimental online study, 225 participants interacted with three fictive classifiers with equal accuracy (73%). The classifiers made prediction mistakes on three different types of data points (easy, difficult, impossible). After the interaction, participants judged the classifier's accuracy. We found that not all prediction mistakes reduced the perceived accuracy equally. Furthermore, the perceived accuracy differed significantly from the calculated accuracy. To conclude, accuracy and related measures seem unsuitable to represent how users perceive the performance of classifiers. CCS CONCEPTS• Human-centered computing → Empirical studies in HCI; Human computer interaction (HCI).

show abstract

Section: Human-centered Evaluation Of Machine Learning Systemsmentioning

confidence: 99%

Section: Human-centered Evaluation Of Machine Learning Systemsmentioning

confidence: 99%

How Accurate Does It Feel? – Human Perception of Different Types of Classification Mistakes

Papenmeier

Kern

Hienert

et al. 2022

CHI Conference on Human Factors in Computing Systems

View full text Add to dashboard Cite

show abstract

“…Xu et al [72] attempted to investigate children's perception of conversational agents available in smart devices. Smith-Renner et al [73] studied how automatically generated explanations of ML models shape users' perceptions of ML models. Völkel et al [74] explored how to mislead chatbots in profiling users.…”

Section: User Studiesmentioning

confidence: 99%

“…Features of AI models addressing the concerns of users to improve the usability and adoptability of AI systems such as explainability, interpretability, privacy, and fairness have been the focus of many HCML related work [6,12,20,26,58,73,76,81,104,178,219]. This is not surprising, given the history of XAI research area dates back to 1980s [220,221].…”

Section: Features Of the Modelsmentioning

confidence: 99%

A Review of Recent Deep Learning Approaches in Human-Centered Machine Learning

Kaluarachchi

Reis

Nanayakkara

2021

Sensors

View full text Add to dashboard Cite

After Deep Learning (DL) regained popularity recently, the Artificial Intelligence (AI) or Machine Learning (ML) field is undergoing rapid growth concerning research and real-world application development. Deep Learning has generated complexities in algorithms, and researchers and users have raised concerns regarding the usability and adoptability of Deep Learning systems. These concerns, coupled with the increasing human-AI interactions, have created the emerging field that is Human-Centered Machine Learning (HCML). We present this review paper as an overview and analysis of existing work in HCML related to DL. Firstly, we collaborated with field domain experts to develop a working definition for HCML. Secondly, through a systematic literature review, we analyze and classify 162 publications that fall within HCML. Our classification is based on aspects including contribution type, application area, and focused human categories. Finally, we analyze the topology of the HCML landscape by identifying research gaps, highlighting conflicting interpretations, addressing current challenges, and presenting future HCML research opportunities.

show abstract

“…Finally, producing transcripts alongside translations may be framed as producing an explanation (the transcript) alongside the main output (the translation). Research on explainable machine learning systems (Smith-Renner et al, 2020, and references therein) may shed light on desirable properties of these explanation from a usability point of view, as well as questions related to appropriate user interface design.…”

Section: Consistency Vs Accuracymentioning

confidence: 99%

Consistent Transcription and Translation of Speech

Sperber

Setiawan

Gollan

et al. 2020

Transactions of the Association for Computational Linguistics

View full text Add to dashboard Cite

The conventional paradigm in speech translation starts with a speech recognition step to generate transcripts, followed by a translation step with the automatic transcripts as input. To address various shortcomings of this paradigm, recent work explores end-to-end trainable direct models that translate without transcribing. However, transcripts can be an indispensable output in practical applications, which often display transcripts alongside the translations to users. We make this common requirement explicit and explore the task of jointly transcribing and translating speech. Although high accuracy of transcript and translation are crucial, even highly accurate systems can suffer from inconsistencies between both outputs that degrade the user experience. We introduce a methodology to evaluate consistency and compare several modeling approaches, including the traditional cascaded approach and end-to-end models. We find that direct models are poorly suited to the joint transcription/translation task, but that end-to-end models that feature a coupled inference procedure are able to achieve strong consistency. We further introduce simple techniques for directly optimizing for consistency, and analyze the resulting trade-offs between consistency, transcription accuracy, and translation accuracy. 1

show abstract

No Explainability without Accountability

Cited by 81 publications

References 25 publications

How Accurate Does It Feel? – Human Perception of Different Types of Classification Mistakes

How Accurate Does It Feel? – Human Perception of Different Types of Classification Mistakes

A Review of Recent Deep Learning Approaches in Human-Centered Machine Learning

Consistent Transcription and Translation of Speech

Contact Info

Product

Resources

About