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

The Softmax bottleneck was first identified in language modeling as atheoretical limit on the expressivity of Softmax-based models. Being one of themost widely-used methods to output probability, Softmax-based models have founda wide range of applications, including session-based recommender systems(SBRSs). Softmax-based models consist of a Softmax function on top of a finallinear layer. The bottleneck has been shown to be caused by rank deficiency inthe final linear layer due to its connection with matrix factorization. In thispaper, we show that there are more aspects to the Softmax bottleneck in SBRSs.Contrary to common beliefs, overfitting does happen in the final linear layer,while it is often associated with complex networks. Furthermore, we identifiedthat the common technique of sharing item embeddings among session sequencesand the candidate pool creates a tight-coupling that also contributes to thebottleneck. We propose a simple yet effective method, Dropout and Decoupling(D&D), to alleviate these problems. Our experiments show that our methodsignificantly improves the accuracy of a variety of Softmax-based SBRSalgorithms. When compared to other computationally expensive methods, such asMLP and MoS (Mixture of Softmaxes), our method performs on par with and attimes even better than those methods, while keeping the same time complexity asSoftmax-based models.

Breaking the Softmax Bottleneck for Sequential Recommender Systems with Dropout and Decoupling