Scalable Cross-Entropy Loss for Sequential Recommendations with Large Item Catalogs

• URL: http://arxiv.org/abs/2409.18721v1
• Date: Fri, 27 Sep 2024 13:17:59 GMT
• Title: Scalable Cross-Entropy Loss for Sequential Recommendations with Large Item Catalogs
• Authors: Gleb Mezentsev, Danil Gusak, Ivan Oseledets, Evgeny Frolov,
• Abstract summary: This paper introduces a novel Scalable Cross-Entropy (SCE) loss function in the sequential learning setup. It approximates the CE loss for datasets with large-size catalogs, enhancing both time efficiency and memory usage without compromising recommendations quality. Experimental results on multiple datasets demonstrate the effectiveness of SCE in reducing peak memory usage by a factor of up to 100 compared to the alternatives.
• Score: 4.165917157093442
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Scalability issue plays a crucial role in productionizing modern recommender systems. Even lightweight architectures may suffer from high computational overload due to intermediate calculations, limiting their practicality in real-world applications. Specifically, applying full Cross-Entropy (CE) loss often yields state-of-the-art performance in terms of recommendations quality. Still, it suffers from excessive GPU memory utilization when dealing with large item catalogs. This paper introduces a novel Scalable Cross-Entropy (SCE) loss function in the sequential learning setup. It approximates the CE loss for datasets with large-size catalogs, enhancing both time efficiency and memory usage without compromising recommendations quality. Unlike traditional negative sampling methods, our approach utilizes a selective GPU-efficient computation strategy, focusing on the most informative elements of the catalog, particularly those most likely to be false positives. This is achieved by approximating the softmax distribution over a subset of the model outputs through the maximum inner product search. Experimental results on multiple datasets demonstrate the effectiveness of SCE in reducing peak memory usage by a factor of up to 100 compared to the alternatives, retaining or even exceeding their metrics values. The proposed approach also opens new perspectives for large-scale developments in different domains, such as large language models.

Related papers

• Adaptable Embeddings Network (AEN) [49.1574468325115]
  We introduce Adaptable Embeddings Networks (AEN), a novel dual-encoder architecture using Kernel Density Estimation (KDE) AEN allows for runtime adaptation of classification criteria without retraining and is non-autoregressive. The architecture's ability to preprocess and cache condition embeddings makes it ideal for edge computing applications and real-time monitoring systems.
  arXiv  Detail & Related papers  (2024-11-21T02:15:52Z)
• RECE: Reduced Cross-Entropy Loss for Large-Catalogue Sequential Recommenders [4.165917157093442]
  This paper introduces a novel RECE (REduced Cross-Entropy) loss. RECE significantly reduces memory consumption while allowing one to enjoy the state-of-the-art performance of full CE loss. Experimental results on various datasets show that RECE cuts training peak memory usage by up to 12 times compared to existing methods.
  arXiv  Detail & Related papers  (2024-08-05T10:02:29Z)
• ThinK: Thinner Key Cache by Query-Driven Pruning [63.13363917871414]
  Large Language Models (LLMs) have revolutionized the field of natural language processing, achieving unprecedented performance across a variety of applications. This paper focuses on the long-context scenario, addressing the inefficiencies in KV cache memory consumption during inference. We propose ThinK, a novel query-dependent KV cache pruning method designed to minimize attention weight loss while selectively pruning the least significant channels.
  arXiv  Detail & Related papers  (2024-07-30T17:59:08Z)
• Zooming Out on Zooming In: Advancing Super-Resolution for Remote Sensing [31.409817016287704]
  Super-Resolution for remote sensing has the potential for huge impact on planet monitoring. Despite a lot of attention, several inconsistencies and challenges have prevented it from being deployed in practice. This work presents a new metric for super-resolution, CLIPScore, that corresponds far better with human judgments than previous metrics.
  arXiv  Detail & Related papers  (2023-11-29T21:06:45Z)
• ECoFLaP: Efficient Coarse-to-Fine Layer-Wise Pruning for Vision-Language Models [70.45441031021291]
  Large Vision-Language Models (LVLMs) can understand the world comprehensively by integrating rich information from different modalities. LVLMs are often problematic due to their massive computational/energy costs and carbon consumption. We propose Efficient Coarse-to-Fine LayerWise Pruning (ECoFLaP), a two-stage coarse-to-fine weight pruning approach for LVLMs.
  arXiv  Detail & Related papers  (2023-10-04T17:34:00Z)
• Improved Distribution Matching for Dataset Condensation [91.55972945798531]
  We propose a novel dataset condensation method based on distribution matching. Our simple yet effective method outperforms most previous optimization-oriented methods with much fewer computational resources.
  arXiv  Detail & Related papers  (2023-07-19T04:07:33Z)
• Winner-Take-All Column Row Sampling for Memory Efficient Adaptation of Language Model [89.8764435351222]
  We propose a new family of unbiased estimators called WTA-CRS, for matrix production with reduced variance. Our work provides both theoretical and experimental evidence that, in the context of tuning transformers, our proposed estimators exhibit lower variance compared to existing ones.
  arXiv  Detail & Related papers  (2023-05-24T15:52:08Z)
• Efficient and Scalable Recommendation via Item-Item Graph Partitioning [10.390315462253726]
  Collaborative filtering (CF) is a widely searched problem in recommender systems. We propose an efficient and scalable recommendation via item-item graph partitioning (ERGP)
  arXiv  Detail & Related papers  (2022-07-13T04:37:48Z)
• DCT-Former: Efficient Self-Attention with Discrete Cosine Transform [4.622165486890318]
  An intrinsic limitation of the Trasformer architectures arises from the computation of the dot-product attention. Our idea takes inspiration from the world of lossy data compression (such as the JPEG algorithm) to derive an approximation of the attention module. An extensive section of experiments shows that our method takes up less memory for the same performance, while also drastically reducing inference time.
  arXiv  Detail & Related papers  (2022-03-02T15:25:27Z)
• A Generic Network Compression Framework for Sequential Recommender Systems [71.81962915192022]
  Sequential recommender systems (SRS) have become the key technology in capturing user's dynamic interests and generating high-quality recommendations. We propose a compressed sequential recommendation framework, termed as CpRec, where two generic model shrinking techniques are employed. By the extensive ablation studies, we demonstrate that the proposed CpRec can achieve up to 4$sim$8 times compression rates in real-world SRS datasets.
  arXiv  Detail & Related papers  (2020-04-21T08:40:55Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.