Timezone: »

 
Spotlight
One for One, or All for All: Equilibria and Optimality of Collaboration in Federated Learning
Avrim Blum · Nika Haghtalab · Richard Lanas Phillips · Han Shao

Wed Jul 21 07:45 AM -- 07:50 AM (PDT) @
in Learning Theory 6 »

In recent years, federated learning has been embraced as an approach for bringing about collaboration across large populations of learning agents. However, little is known about how collaboration protocols should take agents' incentives into account when allocating individual resources for communal learning in order to maintain such collaborations. Inspired by game theoretic notions, this paper introduces a framework for incentive-aware learning and data sharing in federated learning. Our stable and envy-free equilibria capture notions of collaboration in the presence of agents interested in meeting their learning objectives while keeping their own sample collection burden low. For example, in an envy-free equilibrium, no agent would wish to swap their sampling burden with any other agent and in a stable equilibrium, no agent would wish to unilaterally reduce their sampling burden.

In addition to formalizing this framework, our contributions include characterizing the structural properties of such equilibria, proving when they exist, and showing how they can be computed. Furthermore, we compare the sample complexity of incentive-aware collaboration with that of optimal collaboration when one ignores agents' incentives.

Keywords: Game Theory and Computational Economics

Author Information

Avrim Blum (Toyota Technological Institute of Chicago)
Nika Haghtalab (University of California, Berkeley)
Richard Lanas Phillips (Cornell University)
Han Shao (Toyota Technological Institute at Chicago)

Related Events (a corresponding poster, oral, or spotlight)

More from the Same Authors