Multi-Gradient Descent for Multi-Objective Recommender Systems

Nikola Milojkovic, Diego Antognini, Giancarlo Bergamin, Boi Faltings, Claudiu Musat

Abstract:

Recommender systems need to mirror the complexity of the
environment they are applied in. The more we know about
what might benefit the user, the more objectives the recom-
mender system has. In addition there may be multiple stake-
holders - sellers, buyers, shareholders - in addition to legal
and ethical constraints. Simultaneously optimizing for a mul-
titude of objectives, correlated and not correlated, having the
same scale or not, has proven difficult so far.
We introduce a stochastic multi-gradient descent approach
to recommender systems (MGDRec) to solve this problem.
We show that this exceeds state-of-the-art methods in tradi-
tional objective mixtures, like revenue and recall. Not only
that, but through gradient normalization we can combine fun-
damentally different objectives, having diverse scales, into a
single coherent framework. We show that uncorrelated ob-
jectives, like the proportion of quality products, can be im-
proved alongside accuracy. Through the use of stochasticity,
we avoid the pitfalls of calculating full gradients and provide
a clear setting for its applicability.