Tensions Between the Proxies of Human Values in AI
Dec 2, 2022
Speakers
Teresa Datta
Speaker · 0 followers
Daniel Nissani
Speaker · 0 followers
Max Cembalest
Speaker · 0 followers
Akash Khanna
Speaker · 0 followers
Haley Massa
Speaker · 0 followers
John Dickerson
Speaker · 0 followers
About
Motivated by mitigating potentially harmful impacts of technologies, the AI community has formulated and accepted mathematical definitions for certain pillars of accountability: e.g. privacy, fairness, and model transparency. Yet, we argue this is fundamentally misguided because these definitions are imperfect, siloed constructions of the human values they hope to proxy, while giving the guise that those values are sufficiently embedded in our technologies. Under popularized methods, tensions arise when practitioners attempt to achieve each pillar of fairness, privacy, and transparency in isolation or simultaneously. In this paper, we push for redirection. We argue that the AI community needs to consider alternative formulations of these pillars based on the context in which technology is situated. By leaning on sociotechnical systems research, we can formulate more compatible, domain-specific definitions of our human values for building more ethical systems.Motivated by mitigating potentially harmful impacts of technologies, the AI community has formulated and accepted mathematical definitions for certain pillars of accountability: e.g. privacy, fairness, and model transparency. Yet, we argue this is fundamentally misguided because these definitions are imperfect, siloed constructions of the human values they hope to proxy, while giving the guise that those values are sufficiently embedded in our technologies. Under popularized methods, tensions ar…
Organizer
NeurIPS 2022
Account · 963 followers
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
UMIX: Improving Importance Weighting for Subpopulation Shift via Uncertainty-Aware Mixup
Watch later
Zongbo Han, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Featured Papers Panels 2C
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Transformer-based normative modelling for anomaly detection of early schizophrenia
Watch later
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Safety Guarantees for Neural Network Dynamic Systems via Stochastic Barrier Functions
Watch later
Rayan Mazouz, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Instance-optimal PAC Algorithms for Contextual Bandits
Watch later
Zhaoqi Li, …
Total of 0 viewers voted for saving the presentation to eternal vault which is 0.0%
Lossless Compression of Deep Neural Networks: A High-dimensional Neural Tangent Kernel Approach
Watch later
Lingyu Gu, …
Total of 1 viewers voted for saving the presentation to eternal vault which is 0.1%