Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Promoting User Data Autonomy During the Dissolution of a Monopolistic Firm

Rushabh Solanki, Elliot Creager

TL;DR

It is shown how the framework of Conscious Data Contribution can enable user autonomy during under dissolution and how fine-tuning and the phenomenon of "catastrophic forgetting" could actually prove beneficial as a type of machine unlearning that allows users to specify which data they want used for what purposes.

Abstract

The deployment of AI in consumer products is currently focused on the use of so-called foundation models, large neural networks pre-trained on massive corpora of digital records. This emphasis on scaling up datasets and pre-training computation raises the risk of further consolidating the industry, and enabling monopolistic (or oligopolistic) behavior. Judges and regulators seeking to improve market competition may employ various remedies. This paper explores dissolution -- the breaking up of a monopolistic entity into smaller firms -- as one such remedy, focusing in particular on the technical challenges and opportunities involved in the breaking up of large models and datasets. We show how the framework of Conscious Data Contribution can enable user autonomy during under dissolution. Through a simulation study, we explore how fine-tuning and the phenomenon of "catastrophic forgetting" could actually prove beneficial as a type of machine unlearning that allows users to specify which data they want used for what purposes.

Promoting User Data Autonomy During the Dissolution of a Monopolistic Firm

TL;DR

It is shown how the framework of Conscious Data Contribution can enable user autonomy during under dissolution and how fine-tuning and the phenomenon of "catastrophic forgetting" could actually prove beneficial as a type of machine unlearning that allows users to specify which data they want used for what purposes.

Abstract

The deployment of AI in consumer products is currently focused on the use of so-called foundation models, large neural networks pre-trained on massive corpora of digital records. This emphasis on scaling up datasets and pre-training computation raises the risk of further consolidating the industry, and enabling monopolistic (or oligopolistic) behavior. Judges and regulators seeking to improve market competition may employ various remedies. This paper explores dissolution -- the breaking up of a monopolistic entity into smaller firms -- as one such remedy, focusing in particular on the technical challenges and opportunities involved in the breaking up of large models and datasets. We show how the framework of Conscious Data Contribution can enable user autonomy during under dissolution. Through a simulation study, we explore how fine-tuning and the phenomenon of "catastrophic forgetting" could actually prove beneficial as a type of machine unlearning that allows users to specify which data they want used for what purposes.

Paper Structure

This paper contains 15 sections, 3 equations, 1 figure, 3 tables.

Table of Contents

  1. Introduction
  2. Background
  3. Data Autonomy
  4. Model Adaptation
  5. AI and Market Competition
  6. Conscious Data Contribution Under Firm Dissolution
  7. Simulation Studies
  8. Datasets
  9. Image generation
  10. Text classification
  11. Results
  12. Conclusion
  13. Experimental Details
  14. Image generation
  15. Text classification

Figures (1)

  • Figure 1: The plots illustrates the changes in forget and retain rates during the process of fine-tuning and retraining with consciously contributed data $U^{\text{CDC}}$ to a successor $S$ of a firm upon dissolution. Each plot displays the results for different datasets, with the top two for image generation and bottom one for text classification.