Impossibility of Collective Intelligence

06/05/2022
by   Krikamol Muandet, et al.
0

Democratization of AI involves training and deploying machine learning models across heterogeneous and potentially massive environments. Diversity of data opens up a number of possibilities to advance AI systems, but also introduces pressing concerns such as privacy, security, and equity that require special attention. This work shows that it is theoretically impossible to design a rational learning algorithm that has the ability to successfully learn across heterogeneous environments, which we decoratively call collective intelligence (CI). By representing learning algorithms as choice correspondences over a hypothesis space, we are able to axiomatize them with essential properties. Unfortunately, the only feasible algorithm compatible with all of the axioms is the standard empirical risk minimization (ERM) which learns arbitrarily from a single environment. Our impossibility result reveals informational incomparability between environments as one of the foremost obstacles for researchers who design novel algorithms that learn from multiple environments, which sheds light on prerequisites for success in critical areas of machine learning such as out-of-distribution generalization, federated learning, algorithmic fairness, and multi-modal learning.

READ FULL TEXT

page 4

page 9

research
06/17/2021

Towards Heterogeneous Clients with Elastic Federated Learning

Federated learning involves training machine learning models over device...
research
02/08/2023

Algorithmic Collective Action in Machine Learning

We initiate a principled study of algorithmic collective action on digit...
research
09/17/2021

Enforcing fairness in private federated learning via the modified method of differential multipliers

Federated learning with differential privacy, or private federated learn...
research
02/16/2021

Scaling Neuroscience Research using Federated Learning

The amount of biomedical data continues to grow rapidly. However, the ab...
research
09/18/2023

Context is Environment

Two lines of work are taking the central stage in AI research. On the on...
research
08/25/2020

Accelerating Federated Learning in Heterogeneous Data and Computational Environments

There are situations where data relevant to a machine learning problem a...

Please sign up or login with your details

Forgot password? Click here to reset