A Review of Speech-centric Trustworthy Machine Learning: Privacy, Safety, and Fairness
نویسندگان
چکیده
Speech-centric machine learning systems have revolutionized many leading domains ranging from transportation and healthcare to education defense, profoundly changing how people live, work, interact with each other. However, recent studies demonstrated that speech-centric ML may need be considered more trustworthy for broader deployment. Specifically, concerns over privacy breaches, discriminating performance, vulnerability adversarial attacks all been discovered in research fields. In order address the above challenges risks, a significant number of efforts made ensure these are trustworthy, especially private, safe, fair. this paper, we conduct first comprehensive survey on topics related privacy, safety, fairness. addition serving as summary report community, point out several promising future directions inspire researchers who wish explore further area.
منابع مشابه
Differential Privacy and Machine Learning: a Survey and Review
The objective of machine learning is to extract useful information from data, while privacy is preserved by concealing information. Thus it seems hard to reconcile these competing interests. However, they frequently must be balanced when mining sensitive data. For example, medical research represents an important application where it is necessary both to extract useful information and protect p...
متن کاملa comparative pragmatic analysis of the speech act of “disagreement” across english and persian
the speech act of disagreement has been one of the speech acts that has received the least attention in the field of pragmatics. this study investigates the ways power relations, social distance, formality of the context, gender, and language proficiency (for efl learners) influence disagreement and politeness strategies. the participants of the study were 200 male and female native persian s...
15 صفحه اولFairness-aware machine learning: a perspective
Algorithms learned from data are increasingly used for deciding many aspects in our life: from movies we see, to prices we pay, or medicine we get. Yet there is growing evidence that decision making by inappropriately trained algorithms may unintentionally discriminate people. For example, in automated matching of candidate CVs with job descriptions, algorithms may capture and propagate ethnici...
متن کاملTowards Constructing a Trustworthy Internet: Privacy-Aware Transfer of Digital Identity Document in Content Centric Internetworking
Managing digital identity documents with a proper privacy protection is of pivotal importance to construct trustworthy Internet. As far as the amount of digital identities is expanding at an accelerating rate, content-centric model provides administration capabilities of data transfer. We propose an innovative approach and implementation of privacy-aware Content-Centric Internetworking (CCN)-ba...
متن کاملActivity-Centric Email: A Machine Learning Approach
Our use of ordinary desktop applications (such as email, Web, calendars) is often a manifestation of the activities with which we are engaged (Moran, Cozzi, & Farrell 2005). Planning a conference trip involves sending travel expense forms, and visits to airline and hotel sites. Renovating a kitchen involves sketches, product specifications, emails with the architect and spreadsheets for trackin...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: APSIPA transactions on signal and information processing
سال: 2023
ISSN: ['2048-7703']
DOI: https://doi.org/10.1561/116.00000084