no code implementations • 17 Apr 2023 • Khushhall Chandra Mahajan, Aditya Palnitkar, Ameya Raul, Brad Schumitsch
In this paper, we propose a novel method which introduces diversity by modeling the impact of low diversity on user's engagement on individual items, thus being able to account for both diversity and relevance to adjust item scores.
no code implementations • 17 Apr 2023 • Ameya Raul, Amey Porobo Dharwadker, Brad Schumitsch
Learning large-scale industrial recommender system models by fitting them to historical user interaction data makes them vulnerable to conformity bias.
no code implementations • 13 Apr 2023 • Khushhall Chandra Mahajan, Amey Porobo Dharwadker, Romil Shah, Simeng Qu, Gaurav Bang, Brad Schumitsch
We also analyze the value of exploration by defining relevant metrics around user-creator connections and understanding how this helps the overall recommendation pipeline with strong online gains in creator and ecosystem value.