English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Sequential recommender systems (SRS) have become a research hotspot due toits power in modeling user dynamic interests and sequential behavioralpatterns. To maximize model expressive ability, a default choice is to apply alarger and deeper network architecture, which, however, often brings highnetwork latency when generating online recommendations. Naturally, we arguethat compressing the heavy recommendation models into middle- or light- weightneural networks is of great importance for practical production systems. Torealize such a goal, we propose AdaRec, a knowledge distillation (KD) frameworkwhich compresses knowledge of a teacher model into a student model adaptivelyaccording to its recommendation scene by using differentiable NeuralArchitecture Search (NAS). Specifically, we introduce a target-orienteddistillation loss to guide the structure search process for finding the studentnetwork architecture, and a cost-sensitive loss as constraints for model size,which achieves a superior trade-off between recommendation effectiveness andefficiency. In addition, we leverage Earth Mover's Distance (EMD) to realizemany-to-many layer mapping during knowledge distillation, which enables eachintermediate student layer to learn from other intermediate teacher layersadaptively. Extensive experiments on real-world recommendation datasetsdemonstrate that our model achieves competitive or better accuracy with notableinference speedup comparing to strong counterparts, while discovering diverseneural architectures for sequential recommender models under differentrecommendation scenes.

Scene-adaptive Knowledge Distillation for Sequential Recommendation via Differentiable Architecture Search