Zobrazeno 1 - 5
of 5
pro vyhledávání: '"Lichtenberg, Jan Malte"'
An increasing number of media streaming services have expanded their offerings to include entities of multiple content types. For instance, audio streaming services that started by offering music only, now also offer podcasts, merchandise items, and
Externí odkaz:
http://arxiv.org/abs/2408.09168
The issue of popularity bias -- where popular items are disproportionately recommended, overshadowing less popular but potentially relevant items -- remains a significant challenge in recommender systems. Recent advancements have seen the integration
Externí odkaz:
http://arxiv.org/abs/2406.01285
Autor:
Lichtenberg, Jan Malte, Buchholz, Alexander, Di Benedetto, Giuseppe, Ruffini, Matteo, London, Ben
"Clipping" (a.k.a. importance weight truncation) is a widely used variance-reduction technique for counterfactual off-policy estimators. Like other variance-reduction techniques, clipping reduces variance at the cost of increased bias. However, unlik
Externí odkaz:
http://arxiv.org/abs/2309.01120
Autor:
Buchholz, Alexander, Lichtenberg, Jan Malte, Di Benedetto, Giuseppe, Stein, Yannik, Bellini, Vito, Ruffini, Matteo
The Plackett-Luce (PL) model is ubiquitous in learning-to-rank (LTR) because it provides a useful and intuitive probabilistic model for sampling ranked lists. Counterfactual offline evaluation and optimization of ranking metrics are pivotal for using
Externí odkaz:
http://arxiv.org/abs/2205.06024
Autor:
Lichtenberg, Jan Malte, Şimşek, Özgür
Humans and animals solve a difficult problem much more easily when they are presented with a sequence of problems that starts simple and slowly increases in difficulty. We explore this idea in the context of reinforcement learning. Rather than provid
Externí odkaz:
http://arxiv.org/abs/1912.02532