Continuous-Learning Recommendation Engines: Fusing Deep Metric Embeddings with Contextual Bandits at Web Scale
International Journal of Computational and Experimental Science and Engineering · 21.02.2026 · 12 (1)
DOI: 10.22399/ijcesen.4945 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computational and Experimental Science and Engineering, online erschienen am 21.02.2026. Verfasst von Aditya Choudhary. Die Arbeit nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
A novel recommendation system architecture integrates deep metric-learning embeddings with contextual-bandit exploration to address limitations of traditional recommenders. This hybrid design enables personalized content delivery while continuously adapting to evolving user preferences and contexts. The architecture captures semantic relationships between items through high-dimensional embeddings while systematically exploring new options through contextual bandits, creating a balanced approach to the exploitation-exploration dilemma. Implementation features include efficient feature pipeline orchestration, on-device inference capabilities, and robust counterfactual evaluation techniques. Both offline and online evaluations demonstrate significant improvements in click-through rates, cold-start adaptation speed, and recommendation diversity without sacrificing relevance. Responsible deployment patterns including shadow mode operation, fairness audits, and feedback loop dampening ensure the system functions ethically at web scale.
Zitieren
APA
Choudhary, A. (2026). Continuous-Learning Recommendation Engines: Fusing Deep Metric Embeddings with Contextual Bandits at Web Scale. International Journal of Computational and Experimental Science and Engineering, 12(1). https://doi.org/10.22399/ijcesen.4945
Vancouver
Choudhary A. Continuous-Learning Recommendation Engines: Fusing Deep Metric Embeddings with Contextual Bandits at Web Scale. International Journal of Computational and Experimental Science and Engineering. 2026;12(1). doi:10.22399/ijcesen.4945
BibTeX
@article{Choudhary2026,
author = {Choudhary, Aditya},
title = {Continuous-Learning Recommendation Engines: Fusing Deep Metric Embeddings with Contextual Bandits at Web Scale},
journal = {International Journal of Computational and Experimental Science and Engineering},
year = {2026},
volume = {12},
number = {1},
doi = {10.22399/ijcesen.4945}
}
EUVIMED