arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.04375cs.LGcs.AI
收藏

面向稀疏域序列推荐的自适应加权损失

Adaptive Weighted Loss for Sequential Recommendations on Sparse Domains

PhD Scholar University of the Cumberlands(博士学者 哥伦比亚大学);Senior IEEE Member Independent Researcher(IEEE高级会员 独立研究者)

Akshay Mittal, Vinay Venkatesh, Krishna Kandi, Shalini Sudarshan

AI总结:

针对单模型序列推荐在稀疏域效果受限的问题,提出基于领域稀疏度自适应调整损失权重的动态加权损失方法,经理论验证与多数据集实验,在稀疏域关键指标上显著提升且开销极小。

摘要:

单模型序列推荐架构虽具备可扩展性,但在服务稀疏或小众领域的“重度用户”时,效果往往受限。我们此前的研究PinnerFormerLite(轻量钉选前馈模型)通过固定加权损失来优先处理特定领域,以此解决该问题。但这种方法可能并非最优,因为单一统一权重对交互量极少的领域而言可能不足,这类领域的训练信号很容易被庞大的通用数据集稀释。\n 本文提出一种全新的数据驱动方法:具备完备理论基础与大量实证验证的动态加权损失函数。我们引入一种自适应算法,根据各领域在训练数据中的稀疏度调整其损失权重,为更稀疏的领域分配更高权重,为更密集的领域分配更低权重。这确保了即使用户兴趣十分小众,也能贡献有意义的梯度信号,避免其被掩盖。\n 我们提供了包括收敛性证明、复杂度分析和界分析在内的严格理论分析,以确立本方法的稳定性与高效性。我们在四个不同数据集(MovieLens、Amazon Electronics、Yelp Business、LastFM Music)上,与当前最优基线模型(SIGMA、CALRec、SparseEnNet)开展了全面实证验证。结果表明,该动态加权系统显著优于所有对比方法,尤其在稀疏领域表现突出,在Recall@10和NDCG@10等关键指标上实现了大幅提升,同时保持了密集领域上的性能,且仅引入了极小的计算开销。

英文摘要:

The effectiveness of single-model sequential recommendation architectures, while scalable, is often limited when catering to "power users" in sparse or niche domains. Our previous research, PinnerFormerLite, addressed this by using a fixed weighted loss to prioritize specific domains. However, this approach can be sub-optimal, as a single, uniform weight may not be sufficient for domains with very few interactions, where the training signal is easily diluted by the vast, generic dataset. This paper proposes a novel, data-driven approach: a Dynamic Weighted Loss function with comprehensive theoretical foundations and extensive empirical validation. We introduce an adaptive algorithm that adjusts the loss weight for each domain based on its sparsity in the training data, assigning a higher weight to sparser domains and a lower weight to denser ones. This ensures that even rare user interests contribute a meaningful gradient signal, preventing them from being overshadowed. We provide rigorous theoretical analysis including convergence proofs, complexity analysis, and bounds analysis to establish the stability and efficiency of our approach. Our comprehensive empirical validation across four diverse datasets (MovieLens, Amazon Electronics, Yelp Business, LastFM Music) with state-of-the-art baselines (SIGMA, CALRec, SparseEnNet) demonstrates that this dynamic weighting system significantly outperforms all comparison methods, particularly for sparse domains, achieving substantial lifts in key metrics like Recall at 10 and NDCG at 10 while maintaining performance on denser domains and introducing minimal computational overhead.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑