Learning from Streaming Data when Users Choose
This addresses the challenge of optimizing models in dynamic, user-driven environments for service providers in digital markets, though it appears incremental as it builds on existing feedback loop concepts.
The paper tackles the problem of learning from streaming data in competitive digital markets where user choices and model updates create feedback loops, and presents a decentralized algorithm that asymptotically converges to stationary points of the overall user loss.
In digital markets comprised of many competing services, each user chooses between multiple service providers according to their preferences, and the chosen service makes use of the user data to incrementally improve its model. The service providers' models influence which service the user will choose at the next time step, and the user's choice, in return, influences the model update, leading to a feedback loop. In this paper, we formalize the above dynamics and develop a simple and efficient decentralized algorithm to locally minimize the overall user loss. Theoretically, we show that our algorithm asymptotically converges to stationary points of of the overall loss almost surely. We also experimentally demonstrate the utility of our algorithm with real world data.