A Caltech Library Service

Algorithmic Collective Action in Machine Learning

Hardt, Moritz and Mazumdar, Eric and Mendler-Dünner, Celestine and Zrnic, Tijana (2023) Algorithmic Collective Action in Machine Learning. . (Unpublished)

[img] PDF - Submitted Version
See Usage Policy.


Use this Persistent URL to link to this item:


We initiate a principled study of algorithmic collective action on digital platforms that deploy machine learning algorithms. We propose a simple theoretical model of a collective interacting with a firm's learning algorithm. The collective pools the data of participating individuals and executes an algorithmic strategy by instructing participants how to modify their own data to achieve a collective goal. We investigate the consequences of this model in three fundamental learning-theoretic settings: the case of a nonparametric optimal learning algorithm, a parametric risk minimizer, and gradient-based optimization. In each setting, we come up with coordinated algorithmic strategies and characterize natural success criteria as a function of the collective's size. Complementing our theory, we conduct systematic experiments on a skill classification task involving tens of thousands of resumes from a gig platform for freelancers. Through more than two thousand model training runs of a BERT-like language model, we see a striking correspondence emerge between our empirical observations and the predictions made by our theory. Taken together, our theory and experiments broadly support the conclusion that algorithmic collectives of exceedingly small fractional size can exert significant control over a platform's learning algorithm.

Item Type:Report or Paper (Discussion Paper)
Related URLs:
URLURL TypeDescription Paper
Mazumdar, Eric0000-0002-1815-269X
Mendler-Dünner, Celestine0000-0002-9880-7173
Additional Information:The authors would like to thank Solon Barocas for pointers to related work, and the attendees of the 2023 Annual Meeting of the Simons Collaboration on the Theory of Algorithmic Fairness for feedback on the project. We thank Christos Papadimitriou for stimulating discussions about the work. This work was supported by the Tübingen AI Center.
Funding AgencyGrant Number
Record Number:CaltechAUTHORS:20230316-204028845
Persistent URL:
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:120101
Deposited By: George Porter
Deposited On:16 Mar 2023 23:09
Last Modified:16 Mar 2023 23:09

Repository Staff Only: item control page